INDEX

Explanations

names with the initial "L" along with some associated words or phrases

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'L' followed by a number

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Satanic

-0.67

 âĶľ

-0.67

ãĥ¼ãĥĨãĤ£

-0.66

smanship

-0.64

err

-0.64

 Prohibition

-0.64

 beware

-0.62

 mileage

-0.62

 prohibition

-0.61

 actionGroup

-0.61

POSITIVE LOGITS

ibrarian

1.27

ifestyle

1.22

ocker

1.20

arkin

1.20

aptop

1.14

aughlin

1.13

acey

1.12

ayers

1.12

aund

1.12

ateral

1.11

Activations Density 0.027%

names with the initial "L" along with some associated words or phrases

occurrences of the letter 'L' followed by a number

No Comments

No Known Activations