INDEX

Explanations

mentions of the word "Form" in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 railing

-0.70

>>\

-0.68

EStreamFrame

-0.63

visor

-0.62

sung

-0.62

 spree

-0.59

 Throne

-0.59

foreseen

-0.57

 Leone

-0.56

 Sons

-0.56

POSITIVE LOGITS

aldehyde

1.59

idable

1.34

atter

1.27

atted

1.26

ulating

1.23

ulas

1.16

ative

1.15

ulates

1.13

ulations

1.12

ula

1.11

Activations Density 0.026%

mentions of the word "Form" in various contexts

No Comments

No Known Activations