INDEX

Explanations

references to the frequency or rate of utilization

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

sonian

-0.67

inosaur

-0.66

tell

-0.66

abella

-0.65

hair

-0.60

otics

-0.59

 Kore

-0.59

bitious

-0.59

ness

-0.58

oward

-0.58

POSITIVE LOGITS

 habits

1.10

 patterns

0.88

iences

0.76

 rates

0.75

ername

0.73

Rate

0.71

rate

0.70

 behaviours

0.70

notations

0.70

 trends

0.69

Activations Density 0.036%

references to the frequency or rate of utilization

No Comments

No Known Activations