INDEX

Explanations

names or references to specific locations

oai_token-act-pair · gpt-3.5-turbo

references to specific locations and associated events related to political or social issues

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

ertodd

-0.84

orem

-0.77

redit

-0.68

ITH

-0.66

luster

-0.66

Ĥª

-0.66

rint

-0.64

swing

-0.63

matic

-0.63

 Wyoming

-0.61

POSITIVE LOGITS

 Mubarak

0.84

asser

0.82

Ã©e

0.82

mone

0.81

============

0.79

xual

0.78

éĹĺ

0.76

anca

0.73

Ara

0.73

ousse

0.72

Activations Density 0.042%

names or references to specific locations

references to specific locations and associated events related to political or social issues

No Comments

No Known Activations