INDEX

Explanations

proper nouns or names of individuals, specific places, or entities

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Scots

-0.70

cers

-0.69

 foremost

-0.66

Interstitial

-0.64

riott

-0.64

croft

-0.64

tenance

-0.61

 Staples

-0.61

ACTED

-0.60

 Paulo

-0.59

POSITIVE LOGITS

ombie

1.44

ebra

1.30

ombies

1.30

odiac

1.18

hao

1.07

iggurat

1.05

ymes

1.04

ealous

1.02

ERO

1.02

arro

0.97

Activations Density 1.553%

proper nouns or names of individuals, specific places, or entities

No Comments

No Known Activations