INDEX

Explanations

references to technical information or code snippets containing specific characters

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

etheless

-0.68

iatrics

-0.61

pillar

-0.59

 tense

-0.58

termin

-0.57

 Pastor

-0.56

 sterling

-0.56

 princ

-0.56

STER

-0.56

 endowed

-0.56

POSITIVE LOGITS

adow

1.04

ield

1.04

ooter

1.01

irts

0.98

uffle

0.96

ooters

0.94

apers

0.92

agin

0.90

aders

0.86

ould

0.85

Activations Density 0.027%

references to technical information or code snippets containing specific characters

No Comments

No Known Activations