INDEX

Explanations

mentions of legal or political figures

oai_token-act-pair · gpt-3.5-turbo

non-functional or low-context phrases in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 eleph

-0.82

 mutually

-0.81

 disciplines

-0.77

 synerg

-0.77

 actively

-0.76

 symb

-0.76

 reflex

-0.76

 endeav

-0.76

 favor

-0.75

 casc

-0.75

POSITIVE LOGITS

Copyright

1.52

They

1.44

Advertisement

1.41

Officials

1.39

Topics

1.38

Sources

1.37

It

1.35

According

1.35

Both

1.35

Newsletter

1.33

Activations Density 0.561%

mentions of legal or political figures

non-functional or low-context phrases in a text

No Comments

No Known Activations