INDEX

Explanations

words related to criticism or judgment

oai_token-act-pair · gpt-3.5-turbo

phrases that express different types or kinds of things

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

dn

-0.89

gor

-0.79

ĸļ

-0.77

UF

-0.77

idelines

-0.76

alf

-0.75

Ð²

-0.75

ï¸

-0.74

none

-0.74

APS

-0.73

POSITIVE LOGITS

 thing

1.56

 stuff

1.12

 behavior

1.05

 crap

1.01

 mischief

0.98

 behaviour

0.98

 shenanigans

0.97

 situation

0.96

 mentality

0.95

 activity

0.93

Activations Density 0.053%

words related to criticism or judgment

phrases that express different types or kinds of things

No Comments

No Known Activations