INDEX

Explanations

conjunctions 'and'

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SPONSORED

-0.71

 WARN

-0.69

 tremend

-0.68

odcast

-0.67

lectic

-0.64

 satell

-0.63

 fasc

-0.62

kefeller

-0.62

 motive

-0.60

 ferment

-0.60

POSITIVE LOGITS

erers

1.15

idate

1.07

erer

1.04

romeda

1.04

rogen

1.00

hra

1.00

ahar

0.94

rea

0.93

ean

0.92

emonium

0.92

Activations Density 0.028%

conjunctions 'and'

No Comments

No Known Activations