INDEX

Explanations

political and ideological terms and concepts

oai_token-act-pair · gpt-3.5-turbo

terms related to political ideologies and social movements

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

 guiName

-0.61

onen

-0.58

é¾į

-0.53

-0.51

 Passage

-0.51

 Piper

-0.51

 Mous

-0.51

lyak

-0.49

 Nich

-0.49

 Kubrick

-0.48

POSITIVE LOGITS

etc

1.25

etc

0.86

âĢ¦)

0.82

whatever

0.78

ect

0.70

âĢ¦

0.61

cheat

0.60

doms

0.59

 âĢ¦

0.59

gal

0.57

Activations Density 0.561%

political and ideological terms and concepts

terms related to political ideologies and social movements

No Comments

No Known Activations