INDEX

Explanations

elements expressing subjective interpretation and personal meaning within narratives or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

IntoConstraints

-0.69

 hearsay

-0.61

stiff

-0.59

ativement

-0.59

principalColumn

-0.54

 Ferri

-0.54

byshev

-0.53

)");

-0.53

raltar

-0.53

DEFINITIONS

-0.52

POSITIVE LOGITS

 hidden

0.69

hidden

0.61

 Hidden

0.61

Hidden

0.59

 discerned

0.53

 discern

0.53

 glean

0.53

 discernible

0.51

 concealed

0.50

 Detected

0.50

Activations Density 0.377%

elements expressing subjective interpretation and personal meaning within narratives or experiences

No Comments

No Known Activations