INDEX

Explanations

dates or years within documents

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

 cumbers

-0.72

FACE

-0.71

 recip

-0.62

LOAD

-0.61

agic

-0.59

hes

-0.59

asher

-0.59

ess

-0.57

 throb

-0.56

oxin

-0.56

POSITIVE LOGITS

-'

0.81

å¹

0.81

rak

0.78

0.77

 onwards

0.76

âĢ²

0.76

#$

0.75

 edition

0.75

obyl

0.74

gress

0.69

Activations Density 0.048%

dates or years within documents

No Comments

No Known Activations