INDEX

Explanations

negative phrases and descriptions related to perception and judgments about people or situations

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

 réception

-0.66

 navidad

-0.54

veu

-0.54

 difíciles

-0.52

 dewasa

-0.52

 taget

-0.52

-0.51

 razón

-0.51

 recepción

-0.51

extérieur

-0.50

POSITIVE LOGITS

rawDesc

1.11

 pure

1.02

HasAnnotation

1.01

✨:

0.97

Tikang

0.94

cyklopedia

0.88

Portale

0.87

AnchorStyles

0.87

setViewportView

0.87

PreferredItem

0.85

Activations Density 0.625%

negative phrases and descriptions related to perception and judgments about people or situations

No Comments

No Known Activations