INDEX

Explanations

references to coercion or compulsion in actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 становника

-0.77

RTLU

-0.76

SpringRunner

-0.76

StoryboardSegue

-0.74

MigrationBuilder

-0.72

 contextLoads

-0.69

SharedCtor

-0.69

 surla

-0.68

yelidikan

-0.68

 Blon

-0.68

POSITIVE LOGITS

 forced

1.52

Forced

1.43

 Forced

1.35

forced

1.31

 forcing

1.16

 compelled

1.14

 compulsory

1.13

 mandatory

1.08

 Compulsory

1.05

 must

1.05

Activations Density 0.460%

references to coercion or compulsion in actions

No Comments

No Known Activations