INDEX

Explanations

references to speakers or speaking roles

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

𝐳

-0.70

ifndef

-0.60

isfile

-0.60

fits

-0.60

('.');

-0.59

jod

-0.59

ribut

-0.58

iarkan

-0.57

-->

-0.57

ned

-0.56

POSITIVE LOGITS

 speaker

3.29

 Speaker

3.08

Speaker

2.99

 speakers

2.95

speaker

2.89

 Speakers

2.79

 SPEAKER

2.69

speakers

2.68

SPEAKER

2.49

Speakers

2.43

Activations Density 0.083%

references to speakers or speaking roles

No Comments

No Known Activations