INDEX

Explanations

terms related to roles in business or industry

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

ffiti

-0.91

hma

-0.85

pread

-0.78

romy

-0.78

equality

-0.77

utenberg

-0.77

erial

-0.76

ori

-0.76

immer

-0.75

ilver

-0.74

POSITIVE LOGITS

 manufacturer

0.79

 subsidiary

0.79

 investor

0.78

 corporation

0.76

 Siem

0.76

 insurer

0.75

 Investor

0.73

 supplier

0.72

's

0.71

 subsidiaries

0.68

Activations Density 0.062%

terms related to roles in business or industry

No Comments

No Known Activations