INDEX

Explanations

references and mentions of the TV station "KAT" in various contexts

oai_token-act-pair · gpt-3.5-turbo

the abbreviation "AT" or context related to media sources

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

perse

-0.85

ppelin

-0.71

produ

-0.66

ingen

-0.66

 Micha

-0.65

doms

-0.64

hold

-0.63

 Janeiro

-0.62

dale

-0.61

 Tsukuyomi

-0.60

POSITIVE LOGITS

ICAN

1.17

RON

1.03

TERN

0.99

ECH

0.96

hetically

0.96

TER

0.95

rix

0.94

oday

0.93

ELY

0.93

itudes

0.93

Activations Density 0.012%

references and mentions of the TV station "KAT" in various contexts

the abbreviation "AT" or context related to media sources

No Comments

No Known Activations