INDEX

Explanations

phrases related to spatial proximity

oai_token-act-pair · gpt-3.5-turbo

references to positions or locations in relation to other items or people

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

ivities

-0.84

Ì¶

-0.69

Includes

-0.68

anguage

-0.67

lance

-0.66

nesota

-0.66

TERN

-0.66

ãĤ¢ãĥ«

-0.66

 understatement

-0.65

adic

-0.65

POSITIVE LOGITS

him

0.97

 hers

0.95

 where

0.94

us

0.91

 ours

0.88

me

0.88

 mine

0.88

 them

0.85

 parked

0.84

rium

0.83

Activations Density 0.166%

phrases related to spatial proximity

references to positions or locations in relation to other items or people

No Comments

No Known Activations