INDEX

Explanations

references to witnesses or eyewitness accounts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 crdi

-0.76

anskje

-0.73

 pleaſure

-0.71

enderror

-0.69

 Exacts

-0.69

 poffible

-0.67

 Moslem

-0.66

 Diſ

-0.65

 >=",

-0.65

moiselle

-0.64

POSITIVE LOGITS

 witness

1.16

 Witness

0.97

witness

0.95

Witness

0.93

 witnessed

0.84

 WITNESS

0.77

 witnessing

0.77

 witnesses

0.73

 neighbor

0.68

 Adam

0.61

Activations Density 0.052%

references to witnesses or eyewitness accounts

No Comments

No Known Activations