INDEX

Explanations

negative adjectives and superlatives

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

 tayo

-0.63

 tungkol

-0.61

ModelMap

-0.54

 vanden

-0.54

 vôtre

-0.51

=$?

-0.51

 kapag

-0.50

 autunno

-0.50

]=="

-0.50

 itong

-0.49

POSITIVE LOGITS

<bos>

0.91

 theolog

0.61

 monaster

0.57

 ideolog

0.57

Uhh

0.56

Diweddarwch

0.56

Xoxo

0.55

Mhm

0.54

°;

0.53

Lma

0.52

Activations Density 0.314%

negative adjectives and superlatives

No Comments

No Known Activations