INDEX

Explanations

occurrences of the word "trip."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

omething

-0.74

anguage

-0.73

IRD

-0.64

 Newfoundland

-0.63

 Canary

-0.62

Liu

-0.62

 linem

-0.60

bodied

-0.59

TOM

-0.59

 Shining

-0.59

POSITIVE LOGITS

ublic

1.05

edia

0.98

athi

0.96

oli

0.96

olitics

0.94

ython

0.91

olic

0.91

athy

0.89

kefeller

0.89

oly

0.88

Activations Density 0.005%

occurrences of the word "trip."

No Comments

No Known Activations