INDEX

Explanations

references to major historical and contemporary conflicts and events

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

requires

-0.80

tips

-0.78

ĳå£«

-0.77

english

-0.76

é¾įåĸļå£«

-0.76

rily

-0.75

verbs

-0.73

uria

-0.73

mobi

-0.73

cedes

-0.72

POSITIVE LOGITS

 debacle

1.52

 fiasco

1.48

 crisis

1.13

 incident

1.13

 saga

1.12

 affair

1.12

 catastrophe

1.11

 scandal

1.10

 situation

1.10

 massacre

1.06

Activations Density 0.173%

references to major historical and contemporary conflicts and events

No Comments

No Known Activations