INDEX

Explanations

names or terms related to specific individuals or locations, particularly those with possible political or geographical significance

oai_token-act-pair · gpt-3.5-turbo

specific political or leadership terms and names

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

levels

-0.68

ngth

-0.66

sylv

-0.65

ĸļ

-0.63

xual

-0.61

 VIDEOS

-0.58

REP

-0.57

 showc

-0.57

olesterol

-0.56

pter

-0.56

POSITIVE LOGITS

oola

0.87

levard

0.83

lehem

0.82

 è£ıè

0.76

abase

0.75

apest

0.72

uda

0.71

oglu

0.69

 Aires

0.69

jamin

0.68

Activations Density 0.231%

names or terms related to specific individuals or locations, particularly those with possible political or geographical significance

specific political or leadership terms and names

No Comments

No Known Activations