INDEX

Explanations

phrases related to companies or organizations

oai_token-act-pair · gpt-3.5-turbo

semicolons in the text

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

ront

-0.82

din

-0.81

senal

-0.81

orts

-0.78

urrection

-0.74

rio

-0.74

hemer

-0.71

izons

-0.70

wald

-0.70

rons

-0.69

POSITIVE LOGITS

 alternatively

1.00

 furthermore

0.97

-)

0.97

 thence

0.97

 moreover

0.88

 consequently

0.85

 namely

0.84

};

0.82

anwhile

0.80

 whereas

0.79

Activations Density 0.054%

phrases related to companies or organizations

semicolons in the text

No Comments

No Known Activations