INDEX

Explanations

words related to a specific historical event or narrative

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 praktik

-0.51

 ditto

-0.51

 depic

-0.51

 excu

-0.49

 konserv

-0.48

 trö

-0.48

 kosme

-0.48

 disgra

-0.47

 shouldnt

-0.47

 pessi

-0.46

POSITIVE LOGITS

⌀

0.50

 Underwear

0.47

ഔ

0.47

StructEnd

0.46

 numerus

0.45

 DELLE

0.45

fjspx

0.44

Witaj

0.44

 totiž

0.44

 broderie

0.43

Activations Density 0.690%

words related to a specific historical event or narrative

No Comments

No Known Activations