INDEX

Explanations

a list of items or rankings

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

 Zwar

-0.69

 umożli

-0.65

Lmfao

-0.65

 Uniwers

-0.60

cze

-0.59

 Leurs

-0.59

 dostar

-0.58

 Gleichzeitig

-0.57

 didel

-0.57

 Pourtant

-0.57

POSITIVE LOGITS

 circon

0.94

 ordina

0.87

 dispen

0.87

 oner

0.86

 dissi

0.86

 embra

0.85

 istan

0.84

 Kategor

0.83

 ciment

0.83

 allarg

0.81

Activations Density 0.119%

a list of items or rankings

No Comments

No Known Activations