INDEX

Explanations

words related to names of people or characters

oai_token-act-pair · gpt-3.5-turbo

words associated with the name "Thad Cochran" and related references

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

lished

-0.76

âĢ¢âĢ¢

-0.75

é¾

-0.66

ãĤ±

-0.65

ãĥĨ

-0.64

ãĤ¦

-0.63

ãĥĩ

-0.63

ENC

-0.62

 kcal

-0.60

Spons

-0.59

POSITIVE LOGITS

issan

0.81

mort

0.80

inge

0.77

acket

0.73

ttle

0.72

iasis

0.69

mology

0.67

bush

0.66

way

0.66

ung

0.66

Activations Density 0.076%

words related to names of people or characters

words associated with the name "Thad Cochran" and related references

No Comments

No Known Activations