INDEX

Explanations

the number "dozen"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "dozen" 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 @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

hra

-0.73

estine

-0.73

claimed

-0.71

 mathemat

-0.70

bro

-0.69

ufact

-0.66

lands

-0.65

ICO

-0.61

otto

-0.61

itter

-0.60

POSITIVE LOGITS

 Thousand

0.98

acity

0.96

 thousand

0.96

uous

0.88

uously

0.87

 dozen

0.85

 consecutive

0.82

 IMAGES

0.80

 iterations

0.77

dozen

0.77

Activations Density 0.018%

the number "dozen"

mentions of the word "dozen" in various contexts

No Comments

No Known Activations