INDEX

Explanations

relevant information related to legal issues and political statements

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

fter

-0.82

ggles

-0.77

][/

-0.75

rome

-0.69

gery

-0.64

tale

-0.63

[/

-0.63

iece

-0.63

eah

-0.63

Ud

-0.63

POSITIVE LOGITS

 itself

0.80

its

0.72

 predecessor

0.69

 reserves

0.68

 annex

0.68

 advert

0.67

 neighbour

0.66

 divisions

0.65

 operational

0.65

0.64

Activations Density 0.375%

relevant information related to legal issues and political statements

No Comments

No Known Activations