INDEX

Explanations

mentions of notable individuals' names

oai_token-act-pair · gpt-3.5-turbo

words related to negative connotations or undesirable situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

pection

-0.66

SPONSORED

-0.65

owship

-0.65

SAY

-0.62

 Takeru

-0.61

 ster

-0.59

 footing

-0.59

cling

-0.58

GBT

-0.58

SPA

-0.57

POSITIVE LOGITS

vous

0.95

ÅĤ

0.87

owicz

0.77

acan

0.76

henko

0.75

ewski

0.72

inis

0.71

Ã¡n

0.70

 Mehran

0.70

adesh

0.69

Activations Density 0.235%

mentions of notable individuals' names

words related to negative connotations or undesirable situations

No Comments

No Known Activations