INDEX

Explanations

phrases with the word "said"

oai_token-act-pair · gpt-3.5-turbo

the repetition of the word "said" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

otype

-0.75

à¦

-0.74

EMBER

-0.72

otin

-0.70

Holy

-0.70

Cop

-0.69

Occup

-0.68

Hung

-0.68

Pont

-0.67

Newsletter

-0.66

POSITIVE LOGITS

 goodbye

0.88

doms

0.79

 anecd

0.73

 afterward

0.72

 emphatically

0.71

 bluntly

0.70

 afterwards

0.69

Fib

0.63

:]

0.63

UFF

0.63

Activations Density 0.229%

phrases with the word "said"

the repetition of the word "said" in various contexts

No Comments

No Known Activations