INDEX

Explanations

famous names, particularly in the context of sports

oai_token-act-pair · gpt-3.5-turbo 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

cock

-0.82

ural

-0.78

abor

-0.73

obar

-0.73

anson

-0.69

ured

-0.68

ax

-0.68

 Grave

-0.68

AX

-0.67

urally

-0.67

POSITIVE LOGITS

 Nunes

0.90

 Reese

0.79

 Staples

0.75

ipolar

0.73

ters

0.71

 Bieber

0.70

dyl

0.69

REDACTED

0.68

ruary

0.68

kson

0.68

Activations Density 0.042%

famous names, particularly in the context of sports

No Comments

No Known Activations