INDEX

Explanations

words associated with identity, recognition, and the consequences of social actions

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

 Nadu

-0.81

 controversies

-0.64

 Quarterly

-0.62

BUG

-0.61

çīĪ

-0.60

 Hawth

-0.60

$$$$

-0.58

Ô

-0.58

Crit

-0.58

 Sabha

-0.58

POSITIVE LOGITS

itely

0.89

iencies

0.85

oppers

0.79

arent

0.78

illet

0.77

icient

0.77

asion

0.76

ades

0.75

igure

0.73

antly

0.73

Activations Density 0.017%

words associated with identity, recognition, and the consequences of social actions

No Comments

No Known Activations