INDEX

Explanations

descriptions of actions or decisions related to political or social issues

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

itiz

-0.65

âĿ

-0.63

hz

-0.60

(?,

-0.59

IDENT

-0.59

eno

-0.58

 Defenders

-0.57

Hot

-0.56

66666666

-0.56

ret

-0.56

POSITIVE LOGITS

 preferably

0.78

ardless

0.77

cause

0.74

 perhaps

0.73

 evidenced

0.72

which

0.71

 whom

0.70

etheless

0.67

nown

0.67

_-

0.67

Activations Density 1.728%

descriptions of actions or decisions related to political or social issues

No Comments

No Known Activations