INDEX

Explanations

terms and phrases related to proof or evidence in a context of argumentation or claims

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

ecycle

-0.84

tones

-0.78

sie

-0.71

ascus

-0.69

asms

-0.66

swer

-0.66

ophon

-0.65

idays

-0.64

bid

-0.63

otin

-0.63

POSITIVE LOGITS

 thereof

0.89

ially

0.73

alled

0.72

lessly

0.71

 proof

0.71

 conclusive

0.70

alling

0.68

 bias

0.68

rence

0.66

 tampering

0.64

Activations Density 0.027%

terms and phrases related to proof or evidence in a context of argumentation or claims

No Comments

No Known Activations