INDEX

Explanations

references to award nominations and wins in the context of film

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

RTEX

-0.54

ковник

-0.47

 Commando

-0.45

zac

-0.44



-0.43

Homepage

-0.42

verti

-0.42

 Load

-0.42

ahon

-0.41

companied

-0.40

POSITIVE LOGITS

 Oscars

0.92

 nominations

0.88

 nomination

0.88

scars

0.86

 awards

0.82

 Oscar

0.79

 nominees

0.79

########.

0.78

nomin

0.77

 nominated

0.76

Activations Density 0.107%

references to award nominations and wins in the context of film

No Comments

No Known Activations