INDEX

Explanations

references to historical events or figures, especially related to war or film history

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

<bos>

-0.79

 flanges

-0.75

 ductile

-0.72

 shewn

-0.69

 exchangers

-0.65

FORMANCE

-0.65

 sprigs

-0.65

 nutella

-0.64

 parenchyma

-0.64

 bituminous

-0.63

POSITIVE LOGITS

 confé

1.27

 Sén

1.14

 vété

1.10

 flé

1.05

 délib

1.02

 clô

0.98

 prédé

0.98

 habile

0.98

 Gouvernement

0.97

 fameux

0.96

Activations Density 0.259%

references to historical events or figures, especially related to war or film history

No Comments

No Known Activations