INDEX

Explanations

mentions of the word "Ma" followed by a numerical value or specific words

oai_token-act-pair · gpt-3.5-turbo

references to the name "Ma" 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 @ 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

graduate

-0.73

tics

-0.71

starter

-0.70

ĳå£«

-0.69

suit

-0.68

ttes

-0.68

itialized

-0.68

lined

-0.67

tyard

-0.67

spection

-0.66

POSITIVE LOGITS

estro

1.27

ureen

1.18

isel

1.01

emon

0.96

uth

0.92

Ma

0.88

iami

0.87

plin

0.86

Na

0.86

ud

0.86

Activations Density 0.010%

mentions of the word "Ma" followed by a numerical value or specific words

references to the name "Ma" in various contexts

No Comments

No Known Activations