INDEX

Explanations

names of geographical locations, specifically focusing on areas in India

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

æĸ¹

-0.64

ONSORED

-0.60

 paddle

-0.60

FUL

-0.59

 judging

-0.58

ATS

-0.57

 semester

-0.57

 Clinic

-0.56

 subp

-0.56

 Simpson

-0.56

POSITIVE LOGITS

 Aviv

1.32

ugu

1.08

stra

1.05

estial

1.04

anyahu

0.95

eno

0.91

angelo

0.89

lez

0.86

cos

0.83

fen

0.82

Activations Density 0.046%

names of geographical locations, specifically focusing on areas in India

No Comments

No Known Activations