INDEX

Explanations

specific names, particularly authors and locations

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

<bos>

-2.22

addComponent

-0.68

 onCancelled

-0.68

imanapun

-0.67

новништво

-0.65

AddTagHelper

-0.65

getSystemService

-0.65

 onResponse

-0.64

enumerate

-0.63

Nhưng

-0.63

POSITIVE LOGITS

 accla

1.62

 affor

1.57

 maneu

1.55

 shenan

1.48

Wtf

1.42

 impra

1.41

 sappi

1.37

 véhic

1.36

 scrat

1.35

 increa

1.35

Activations Density 1.816%

specific names, particularly authors and locations

No Comments

No Known Activations