INDEX

Explanations

words related to demonization or criticism

oai_token-act-pair · gpt-3.5-turbo

references to demonization and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ippi

-0.97

RAFT

-0.79

IGH

-0.74

ļéĨĴ

-0.69

sers

-0.69

Ã¥

-0.68

 Seah

-0.66

aird

-0.65

ILLE

-0.65

rehensive

-0.65

POSITIVE LOGITS

stration

0.94

iac

0.93

ises

0.87

ised

0.86

izing

0.86

ising

0.86

ized

0.82

ization

0.81

oid

0.81

izes

0.81

Activations Density 0.007%

words related to demonization or criticism

references to demonization and related concepts

No Comments

No Known Activations