INDEX

Explanations

words related to directions or locations, specifically focusing on 'north' and 'south'

oai_token-act-pair · gpt-3.5-turbo

conjunctions, particularly the word "and."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.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

tis

-0.84

vu

-0.76

aye

-0.72

manship

-0.72

bo

-0.71

ombies

-0.69

natureconservancy

-0.68

bill

-0.68

conom

-0.67

sbm

-0.65

POSITIVE LOGITS

 periphery

1.10

 rear

0.94

 bottom

0.94

 foremost

0.87

 center

0.83

 forth

0.83

 centre

0.82

 sides

0.81

 outs

0.81

 sidelines

0.80

Activations Density 0.124%

words related to directions or locations, specifically focusing on 'north' and 'south'

conjunctions, particularly the word "and."

No Comments

No Known Activations