INDEX

Explanations

references to the term "son."

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

veyard

-0.79

Union

-0.66

 hazards

-0.62

 manifold

-0.61

Population

-0.60

iculty

-0.59

 prevailing

-0.59

 mechanisms

-0.59

 Tribunal

-0.58

 freight

-0.57

POSITIVE LOGITS

ogram

1.03

nets

0.95

hood

0.94

 Barron

0.91

 Gohan

0.89

son

0.82

pins

0.82

zai

0.81

ograms

0.80

athan

0.79

Activations Density 0.028%

references to the term "son."

No Comments

No Known Activations