INDEX

Explanations

words related to people and their actions or roles in society

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

©¶æ

-0.78

isoft

-0.69

 referen

-0.64

Priv

-0.64

ASED

-0.63

 subp

-0.62

 tandem

-0.60

 insensitive

-0.60

 Expend

-0.59

 divid

-0.58

POSITIVE LOGITS

swick

1.24

shire

1.02

worth

1.01

beck

0.96

inki

0.92

adesh

0.90

fleet

0.89

eston

0.86

iland

0.86

field

0.85

Activations Density 0.061%

words related to people and their actions or roles in society

No Comments

No Known Activations