INDEX

Explanations

phrases containing the word "pin"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "pin" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

anwhile

-0.80

 behav

-0.78

 msec

-0.63

hire

-0.62

 lawy

-0.62

 Emin

-0.60

 Continue

-0.59

 Tribunal

-0.58

Ö¼

-0.58

 plur

-0.58

POSITIVE LOGITS

ning

1.39

ned

1.29

ching

1.17

ochet

1.16

keye

1.15

ched

1.14

ball

1.09

wheel

1.08

hole

1.05

strip

1.02

Activations Density 0.027%

phrases containing the word "pin"

instances of the word "pin" in various contexts

No Comments

No Known Activations