INDEX

Explanations

references to physical locations or objects

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ificant

-0.60

agra

-0.56

avorite

-0.55

 unde

-0.52

 compr

-0.51

 respons

-0.51

pex

-0.51

renheit

-0.50

role

-0.50

dinand

-0.50

POSITIVE LOGITS

):

0.76

 RELEASE

0.67

↵↵

0.58

 standpoint

0.56

':

0.55

Ĥİ

0.55

)?

0.55

↵

0.54

IRE

0.53

photos

0.52

Activations Density 0.570%

references to physical locations or objects

No Comments

No Known Activations