INDEX

Explanations

references to specific proper nouns in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to specific locations and records

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

apsed

-0.89

aser

-0.85

uated

-0.80

ĻĤ

-0.79

ENTS

-0.79

iors

-0.78

ency

-0.74

uates

-0.74

ulate

-0.72

auer

-0.71

POSITIVE LOGITS

shire

0.98

bred

0.82

forth

0.81

ãĤ¦ãĤ¹

0.78

 Ness

0.77

 Farage

0.76

ortment

0.74

comings

0.74

mania

0.70

UTERS

0.68

Activations Density 0.046%

references to specific proper nouns in various contexts

references to specific locations and records

No Comments

No Known Activations