INDEX

Explanations

words related to specific events, actions, or characteristics

oai_token-act-pair · gpt-3.5-turbo

keywords related to competitive situations or conflicts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

cium

-0.61

AFB

-0.54

 Coat

-0.50

Recommend

-0.48

 immunity

-0.48

 Cycling

-0.47

 Codec

-0.46

 urgent

-0.46

 Transcript

-0.46

 Sachs

-0.45

POSITIVE LOGITS

iest

0.99

liest

0.91

less

0.79

lessly

0.73

lessness

0.72

loving

0.66

same

0.65

atics

0.63

most

0.63

mentioned

0.61

Activations Density 0.638%

words related to specific events, actions, or characteristics

keywords related to competitive situations or conflicts

No Comments

No Known Activations