INDEX

Explanations

references to specific years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

flo

-0.79

itionally

-0.70

 subsequ

-0.67

ierrez

-0.66

ishing

-0.64

ishes

-0.62

abulary

-0.61

bars

-0.60

izers

-0.60

 responsive

-0.60

POSITIVE LOGITS

rd

0.98

-'

0.80

å¹

0.78

 onwards

0.74

bard

0.73

RR

0.67

 edition

0.67

etheus

0.66

enary

0.65

MY

0.65

Activations Density 0.023%

references to specific years

No Comments

No Known Activations