INDEX

Explanations

terms related to history, including historical events, figures, and cultural contexts

oai_token-act-pair · gpt-3.5-turbo

references to historical topics and discussions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

entle

-0.86

mods

-0.82

yip

-0.76

cipled

-0.74

assi

-0.73

BILITIES

-0.72

SpaceEngineers

-0.71

itone

-0.70

amaru

-0.70

zeb

-0.70

POSITIVE LOGITS

 flashbacks

0.96

 mythology

0.96

 prol

0.95

 folklore

0.94

 rewritten

0.94

 geography

0.90

 narratives

0.88

 fiction

0.86

 histor

0.86

 flashback

0.86

Activations Density 0.592%

terms related to history, including historical events, figures, and cultural contexts

references to historical topics and discussions

No Comments

No Known Activations