INDEX

Explanations

words related to misinformation and deceit

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

principalTable

-0.71

LabelTagHelper

-0.58

AxisAlignment

-0.53

ActionCreators

-0.52

󠁢

-0.50

الإنجليزية

-0.49

 Winaray

-0.47

 océ

-0.47

 grammi

-0.46

bootstrapcdn

-0.46

POSITIVE LOGITS

 viewers

1.10

 audiences

1.04

 readers

1.04

 listeners

0.98

readers

0.87

 fans

0.86

 audience

0.85

 visitors

0.81

audience

0.81

 viewer

0.80

Activations Density 0.346%

words related to misinformation and deceit

No Comments

No Known Activations