INDEX

Explanations

the word "or" 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

PDATE

-0.74

 Candidate

-0.69

ETS

-0.68

EMP

-0.67

INS

-0.65

 Redmond

-0.65

ESE

-0.65

ergic

-0.64

BLIC

-0.59

ONSORED

-0.58

POSITIVE LOGITS

chard

1.07

acles

1.07

acular

0.95

acle

0.93

ific

0.88

phan

0.86

thodox

0.85

nam

0.83

anges

0.79

 even

0.78

Activations Density 0.040%

the word "or" in various contexts

No Comments

No Known Activations