INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

references to a specific individual or entity, particularly in contexts relating to media or law enforcement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

rity

-0.73

AQ

-0.71

 Views

-0.66

 Euph

-0.65

ortion

-0.64

lished

-0.64

kward

-0.64

ansas

-0.63

awaru

-0.63

ADRA

-0.63

POSITIVE LOGITS

eman

1.06

ergic

0.96

lin

0.89

emen

0.89

quin

0.88

zer

0.87

endon

0.82

etooth

0.82

 Rouge

0.81

hurst

0.80

Activations Density 0.009%

proper nouns

references to a specific individual or entity, particularly in contexts relating to media or law enforcement

No Comments

No Known Activations