INDEX

Explanations

words indicating simultaneous actions or occurrences

oai_token-act-pair · gpt-4o-mini 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

é¾įå¥ĳå£«

-0.72

 Ever

-0.62

maximum

-0.62

 "$:/

-0.61

assic

-0.61

 EVEN

-0.61

 ILCS

-0.61

even

-0.60

liest

-0.59

onz

-0.59

POSITIVE LOGITS

 affected

0.90

 impacted

0.86

 contributed

0.86

 implicated

0.83

 benefited

0.80

 participated

0.77

 included

0.73

 excluded

0.73

 effected

0.72

ypes

0.72

Activations Density 0.102%

words indicating simultaneous actions or occurrences

No Comments

No Known Activations