INDEX

Explanations

descriptive words related to character emotions and story elements

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Intere

-1.39

Juf

-1.33

 frankfurt

-1.27

 pamph

-1.27

hcm

-1.26

 wien

-1.26

 Confu

-1.26

aen

-1.25

 Bibl

-1.24

 coö

-1.23

POSITIVE LOGITS

<bos>

0.73

Vielleicht

0.69

SourceChecksum

0.66

Leider

0.62

 superb

0.61

cassert

0.60

 censiti

0.60

MessageState

0.59

 excellent

0.59

 StatefulWidget

0.58

Activations Density 0.344%

descriptive words related to character emotions and story elements

No Comments

No Known Activations