INDEX

Explanations

mathematical expressions and formulae involving variables like Ι and related operators

oai_token-act-pair · gpt-3.5-turbo

symbols or characters within text that may indicate special formatting or encoding

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ividual

-0.70

 Leilan

-0.63

 Soft

-0.63

 Kens

-0.62

anity

-0.62

icity

-0.60

ality

-0.58

 Moder

-0.58

aciously

-0.57

 hats

-0.57

POSITIVE LOGITS

´

1.79

Ń

1.79

¬

1.76

®

1.75

Į

1.74

·

1.73

¸

1.72

£

1.65

²

1.64

¶

1.62

Activations Density 0.020%

mathematical expressions and formulae involving variables like Ι and related operators

symbols or characters within text that may indicate special formatting or encoding

No Comments

No Known Activations