INDEX

Explanations

technical specifications and components used in various tools or systems

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

stellar

-0.91

FUL

-0.74

theless

-0.73

 acknow

-0.66

)=(

-0.66

 diplom

-0.64

REDACTED

-0.63

é¾įå

-0.59

 Bohem

-0.59

Dar

-0.59

POSITIVE LOGITS

acts

1.08

alyses

1.07

ancers

1.06

rors

1.05

asures

1.03

airs

0.99

ifts

0.98

ippers

0.97

enses

0.96

riers

0.96

Activations Density 1.450%

technical specifications and components used in various tools or systems

No Comments

No Known Activations