INDEX

Explanations

comparisons using the word "Like"

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

teborg

-0.57

 protokol

-0.52

 balkon

-0.52

 alkoh

-0.51

 manuten

-0.51

 keramik

-0.50

pó

-0.49

 asfal

-0.48

jgl

-0.48

szt

-0.48

POSITIVE LOGITS

 intersper

0.83

seperti

0.80

 like

0.80

Like

0.80

ike

0.78

 Like

0.78

like

0.76

LIKE

0.76

 LIKE

0.73

 shenan

0.72

Activations Density 0.092%

comparisons using the word "Like"

No Comments

No Known Activations