INDEX

Explanations

proper nouns related to a specific geographical location, San Francisco

oai_token-act-pair · gpt-3.5-turbo

references to San Francisco

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

mble

-0.89

hift

-0.72

 Beir

-0.72

onut

-0.71

xus

-0.69

ilater

-0.69

iosyncr

-0.68

ividual

-0.67

respons

-0.67

bil

-0.67

POSITIVE LOGITS

Bay

1.08

 Chronicle

0.96

 Francisco

0.95

 Examiner

0.93

 Giants

0.93

 Unified

0.85

 County

0.82

 skyline

0.82

 Rican

0.77

 Municipal

0.74

Activations Density 0.017%

proper nouns related to a specific geographical location, San Francisco

references to San Francisco

No Comments

No Known Activations