INDEX

Explanations

names related to individuals in various contexts

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to individuals and places

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Wellington

-0.82

 Resort

-0.80

pless

-0.79

 Tennis

-0.78

 resort

-0.75

 Pamela

-0.74

 Pizza

-0.73

Alice

-0.70

 Vegan

-0.70

Front

-0.69

POSITIVE LOGITS

 Goodwin

1.76

 Fors

1.56

 symb

1.52

Hod

1.25

 hatch

1.14

 bullish

1.14

hatt

1.07

 Mant

1.05

 mism

1.01

 Targ

0.99

Activations Density 0.084%

names related to individuals in various contexts

proper nouns related to individuals and places

No Comments

No Known Activations