INDEX

Explanations

mentions of specific technical terms or entities, possibly related to a specific field or topic

oai_token-act-pair · gpt-3.5-turbo

terms related to experimental subjects and studies, particularly in the context of a scientific framework

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ources

-0.70

 rooting

-0.65

mouse

-0.61

Reviewed

-0.60

 Krug

-0.60

xual

-0.59

 mileage

-0.58

 advise

-0.57

orescent

-0.57

enhagen

-0.57

POSITIVE LOGITS

etus

0.90

onen

0.85

Lago

0.79

zona

0.77

nova

0.76

atari

0.74

oglu

0.72

ensis

0.69

esi

0.69

ccording

0.67

Activations Density 0.658%

mentions of specific technical terms or entities, possibly related to a specific field or topic

terms related to experimental subjects and studies, particularly in the context of a scientific framework

No Comments

No Known Activations