INDEX

Explanations

words related to political figures or government positions

oai_token-act-pair · gpt-3.5-turbo

words related to local government officials or council members

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

mbuds

-0.66

MU

-0.64

 Feast

-0.63

HP

-0.59

 LIFE

-0.59

PAT

-0.59

EF

-0.58

 chancellor

-0.58

falls

-0.58

 tremend

-0.57

POSITIVE LOGITS

ory

0.98

ors

0.95

orio

0.88

urai

0.88

utes

0.88

ĳ

0.86

arin

0.85

une

0.84

ocol

0.82

aires

0.81

Activations Density 0.028%

words related to political figures or government positions

words related to local government officials or council members

No Comments

No Known Activations