INDEX

Explanations

words related to guided tours

oai_token-act-pair · gpt-3.5-turbo

references to guided tours

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

iod

-0.71

 loose

-0.66

ever

-0.64

 clustered

-0.63

corn

-0.63

 retri

-0.62

 predictive

-0.62

omething

-0.62

ioch

-0.61

pmwiki

-0.61

POSITIVE LOGITS

neys

1.12

 tour

1.06

ney

1.06

 tours

0.97

neau

0.95

 Tours

0.92

 Tour

0.87

efully

0.86

 touring

0.82

icular

0.79

Activations Density 0.018%

words related to guided tours

references to guided tours

No Comments

No Known Activations