INDEX

Explanations

self-reflective statements or emotional responses intertwined with moral judgments

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

mef

-1.66

 lele

-1.66

 haup

-1.63

hcm

-1.61

 wien

-1.57

fta

-1.56

 lyon

-1.52

 embra

-1.52

 squa

-1.52

aen

-1.52

POSITIVE LOGITS

 help

0.73

 give

0.71

 encourage

0.70

 helping

0.68

 tell

0.67

 provide

0.66

<bos>

0.65

 gave

0.64

let

0.63

 teach

0.62

Activations Density 0.634%

self-reflective statements or emotional responses intertwined with moral judgments

No Comments

No Known Activations