INDEX

Explanations

names and terms related to different individuals and locations

oai_token-act-pair · gpt-3.5-turbo

words or phrases associated with specific names or identifiers

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

Marie

-0.77

 Galile

-0.65

 drib

-0.63

ngth

-0.61

inav

-0.59

reditary

-0.59

xual

-0.58

©¶æ

-0.58

uania

-0.57

ources

-0.57

POSITIVE LOGITS

aneers

0.75

istan

0.74

levard

0.73

ylon

0.73

essim

0.71

dit

0.69

ÃĽ

0.69

ecause

0.65

edia

0.65

ournals

0.63

Activations Density 0.270%

names and terms related to different individuals and locations

words or phrases associated with specific names or identifiers

No Comments

No Known Activations