INDEX

Explanations

terms related to familiarity or lack of familiarity with various subjects

oai_token-act-pair · gpt-3.5-turbo

terms related to familiarity and understanding of conditions or concepts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åī

-0.52

utm

-0.51

²¾

-0.49

ampa

-0.48

OUT

-0.47

tein

-0.46

tera

-0.46

wolves

-0.46

°

-0.45

catentry

-0.45

POSITIVE LOGITS

 with

1.25

 withd

1.23

 WITH

1.21

with

1.14

 With

1.07

With

1.02

WithNo

1.00

lished

0.77

icating

0.75

wi

0.69

Activations Density 0.462%

terms related to familiarity or lack of familiarity with various subjects

terms related to familiarity and understanding of conditions or concepts

No Comments

No Known Activations