INDEX

Explanations

words related to action and violence

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

gling

-0.72

orf

-0.63

inately

-0.63

Fey

-0.61

olls

-0.61

 porous

-0.61

owship

-0.60

oiler

-0.60

ringe

-0.60

mbuds

-0.59

POSITIVE LOGITS

ives

0.93

ivism

0.91

ivated

0.89

 Replay

0.86

iveness

0.85

ality

0.80

able

0.80

ual

0.80

aries

0.79

uated

0.77

Activations Density 0.482%

words related to action and violence

No Comments

No Known Activations