INDEX

Explanations

names mentioned within text

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of names in various contexts

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

OPLE

-0.77

UTERS

-0.72

yrinth

-0.71

Constructed

-0.71

irth

-0.70

UGE

-0.69

Returns

-0.66

ETHOD

-0.62

 Springs

-0.62

Decre

-0.62

POSITIVE LOGITS

paces

1.64

pace

1.25

paced

1.11

plates

0.99

ames

0.96

hips

0.90

peed

0.90

erver

0.90

 aliases

0.88

 names

0.87

Activations Density 0.028%

names mentioned within text

the repeated mention of names in various contexts

No Comments

No Known Activations