INDEX

Explanations

references to specific names, such as "Tiger" and "Maddow"

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals or characters involved in notable events or stories

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

beit

-0.83

agers

-0.76

iders

-0.71

 Radiation

-0.63

PDATE

-0.61

 disinfect

-0.60

ufact

-0.60

 Scotia

-0.60

 polarized

-0.60

 magnification

-0.60

POSITIVE LOGITS

psey

0.86

reon

0.83

bies

0.79

pillar

0.77

eas

0.77

avanaugh

0.76

eon

0.75

eem

0.74

rane

0.74

iem

0.74

Activations Density 0.048%

references to specific names, such as "Tiger" and "Maddow"

mentions of specific individuals or characters involved in notable events or stories

No Comments

No Known Activations