INDEX

Explanations

phrases indicating comparison of different entities in terms of qualities

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

 Kün

-0.77

 Simult

-0.65

Välislingid

-0.59

Ueb

-0.58

 lele

-0.58

 meras

-0.57

 bambu

-0.56

 tortas

-0.55

Jä

-0.55

Avez

-0.55

POSITIVE LOGITS

 WALTZ

0.59

not

0.56

 nemici

0.56

 ladri

0.55

setzer

0.55

 giapp

0.54

 nemico

0.52

not

0.52

 disrespect

0.51

 meant

0.50

Activations Density 0.146%

phrases indicating comparison of different entities in terms of qualities

No Comments

No Known Activations