INDEX

Explanations

factoring in different gaming-related terms and specific game titles

oai_token-act-pair · gpt-3.5-turbo

names of video games and related media

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

mble

-0.97

perm

-0.91

rac

-0.88

unal

-0.88

orate

-0.87

admin

-0.87

unc

-0.86

resent

-0.85

nc

-0.84

aca

-0.84

POSITIVE LOGITS

 Lord

0.90

 Thrones

0.88

 Suit

0.87

 Crusade

0.81

 Rapids

0.81

 Sons

0.81

 Dome

0.80

 Islanders

0.80

 Whip

0.79

 Wire

0.79

Activations Density 0.346%

factoring in different gaming-related terms and specific game titles

names of video games and related media

No Comments

No Known Activations