INDEX

Explanations

descriptive adjectives and imagery related to environmental conditions and settings

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

biased

-0.85

osponsors

-0.79

thood

-0.76

mbuds

-0.74

iphany

-0.73

Specific

-0.73

cipled

-0.72

qualified

-0.72

vier

-0.71

cop

-0.70

POSITIVE LOGITS

 confines

1.22

est

1.17

 waters

1.09

ness

1.08

 landscape

1.05

 darkness

1.04

 depths

1.03

 surroundings

1.02

 glare

1.02

 aftermath

1.00

Activations Density 0.114%

descriptive adjectives and imagery related to environmental conditions and settings

No Comments

No Known Activations