INDEX

Explanations

words related to behavioral inclinations or propensities

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "tendency" and its contextual implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

estamp

-0.78

idden

-0.76

mberg

-0.75

aban

-0.74

gur

-0.69

oho

-0.69

ÄŁ

-0.69

riel

-0.69

arer

-0.68

loo

-0.66

POSITIVE LOGITS

 tendency

1.05

 toward

0.99

 towards

0.96

 tendencies

0.90

 propensity

0.82

 predis

0.79

 shift

0.78

 prone

0.71

 behaviour

0.71

 aversion

0.71

Activations Density 0.025%

words related to behavioral inclinations or propensities

mentions of the word "tendency" and its contextual implications

No Comments

No Known Activations