INDEX

Explanations

words related to exerting force or pressure on someone or something

oai_token-act-pair · gpt-3.5-turbo

references to coercion or force applied to individuals

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

uded

-0.65

izoph

-0.58

ILA

-0.57

aming

-0.57

von

-0.56

ording

-0.55

 artif

-0.55

effect

-0.55

 Shap

-0.54

angered

-0.54

POSITIVE LOGITS

 into

1.10

to

1.03

into

0.98

 toward

0.96

 onward

0.95

 onto

0.95

 towards

0.92

 onwards

0.87

 thereto

0.86

To

0.79

Activations Density 0.164%

words related to exerting force or pressure on someone or something

references to coercion or force applied to individuals

No Comments

No Known Activations