INDEX

Explanations

instances of the word "report" and its variations in relation to accountability and disclosure

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

creen

-0.75

ulic

-0.66

arta

-0.66

ierre

-0.65

frey

-0.64

erie

-0.63

 gauge

-0.61

eries

-0.60

enei

-0.60

 econom

-0.57

POSITIVE LOGITS

 misconduct

0.83

 sightings

0.82

 inaccurate

0.79

 inaccur

0.77

 anonymously

0.76

 inacc

0.76

 crimes

0.75

 unwanted

0.74

ulate

0.74

 irregularities

0.72

Activations Density 0.023%

instances of the word "report" and its variations in relation to accountability and disclosure

No Comments

No Known Activations