INDEX

Explanations

number patterns related to dates or statistics

oai_token-act-pair · gpt-3.5-turbo

numerical indicators of importance or ranking

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 è£ıè¦ļéĨĴ

-0.74

ãĥ´ãĤ¡

-0.71

chwitz

-0.71

awaru

-0.68

utra

-0.65

kson

-0.63

owered

-0.63

orem

-0.62

 Hots

-0.62

atis

-0.62

POSITIVE LOGITS

rd

1.02

nd

1.01

Â¯Â¯

0.81

0.80

nces

0.75

mpeg

0.74

ald

0.72

pine

0.71

lake

0.68

arms

0.68

Activations Density 0.042%

number patterns related to dates or statistics

numerical indicators of importance or ranking

No Comments

No Known Activations