INDEX

Explanations

words related to criticism or negative judgment

oai_token-act-pair · gpt-3.5-turbo

phrases indicating negative situations or conflicts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

eter

-0.59

Eld

-0.55

 contact

-0.54

ricting

-0.53

iry

-0.53

 weeds

-0.53

isky

-0.52

 withdrawals

-0.52

hed

-0.52

 bushes

-0.52

POSITIVE LOGITS

 mathemat

0.70

ħĭ

0.70

ãĥĳ

0.66

ccording

0.64

 nodd

0.61

 therap

0.60

 psychiat

0.60

worst

0.59

 unnecess

0.59

¤

0.58

Activations Density 0.471%

words related to criticism or negative judgment

phrases indicating negative situations or conflicts

No Comments

No Known Activations