INDEX

Explanations

references to specific years and data points related to statistics

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

years

oai_token-act-pair · gpt-3.5-turbo

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

yout

-0.76

itton

-0.74

atem

-0.73

 lawy

-0.73

cott

-0.69

CHAT

-0.65

helps

-0.64

fax

-0.64

ype

-0.63

 parap

-0.61

POSITIVE LOGITS

 onwards

0.90

âĢ²

0.84

GDP

0.83

regate

0.80

terms

0.76

-'

0.75

 Census

0.72

âĢĳ

0.69

 Definition

0.66

 census

0.65

Activations Density 0.177%

references to specific years and data points related to statistics

years

No Comments

No Known Activations