INDEX

Explanations

information related to news articles and blog posts, such as details about incidents and personal stories

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 massacres

-0.98

 atrocities

-0.87

 bombings

-0.86

 disinformation

-0.86

corruption

-0.86

 insurgents

-0.85

 insurgency

-0.84

ABORT

-0.83

 fascism

-0.82

 traitors

-0.81

POSITIVE LOGITS

<bos>

16.57

expandindo

2.54

GEBURTSDATUM

2.49

 betweenstory

2.40

Autoritní

2.35

 encomp

2.32

 Administrativna

2.30

 intersper

2.29

تقاوى

2.28

 dispen

2.28

Activations Density 0.956%

information related to news articles and blog posts, such as details about incidents and personal stories

No Comments

No Known Activations