INDEX

Explanations

proper nouns related to political figures or events

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

 Marketable

-0.79

incial

-0.76

_-

-0.76

INAL

-0.72

actionDate

-0.67

²¾

-0.65

ĪĴ

-0.64

«ĺ

-0.64

ormal

-0.64

ħĭ

-0.64

POSITIVE LOGITS

nesday

0.79

reens

0.74

ipeg

0.73

theless

0.68

eret

0.63

 Rost

0.62

lich

0.60

hof

0.60

bucks

0.59

wright

0.59

Activations Density 3.194%

proper nouns related to political figures or events

No Comments

No Known Activations