INDEX

Explanations

terms related to significant changes or transformations

oai_token-act-pair · gpt-3.5-turbo

references to paradigm shifts

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

llah

-0.70

idan

-0.67

hire

-0.67

 Cheong

-0.66

 Nationals

-0.63

bleacher

-0.63

akov

-0.62

ionage

-0.62

nery

-0.61

POSITIVE LOGITS

 paradigm

1.08

igmatic

0.95

 parad

0.89

 shift

0.82

igm

0.75

 Parad

0.73

Ħ¢

0.72

 Shift

0.70

ually

0.69

atically

0.68

Activations Density 0.019%

terms related to significant changes or transformations

references to paradigm shifts

No Comments

No Known Activations