INDEX

Explanations

references to threats and dangers

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

 scolas

-0.75

tituts

-0.69

Absorption

-0.68

hoga

-0.68

☀

-0.65

roxene

-0.63

urator

-0.62

adins

-0.61

oflavin

-0.60

uinal

-0.59

POSITIVE LOGITS

 threat

1.82

 threats

1.72

threat

1.72

 Threat

1.64

 Threats

1.60

Threat

1.57

 threatens

1.45

 threatened

1.43

Threats

1.39

 threaten

1.36

Activations Density 0.085%

references to threats and dangers

No Comments

No Known Activations