INDEX

Explanations

words related to specific geographic locations

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to places and names

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

UME

-0.75

ODUCT

-0.74

 urgently

-0.65

ccording

-0.64

 Surviv

-0.64

 corros

-0.63

 perverse

-0.63

 irresist

-0.63

lopp

-0.63

ADRA

-0.63

POSITIVE LOGITS

stown

1.19

ville

1.09

 Creek

1.03

burg

0.97

sonian

0.94

town

0.93

hurst

0.93

 Valley

0.92

 County

0.90

 Canyon

0.89

Activations Density 0.319%

words related to specific geographic locations

proper nouns related to places and names

No Comments

No Known Activations