INDEX

Explanations

negations and expressions of uncertainty or denial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

います

-0.77

 itſelf

-0.76

rungsseite

-0.76

 habet

-0.75

 Majefty

-0.70

 Monfieur

-0.69

 retires

-0.69

 hangs

-0.68

 trône

-0.68

 eorum

-0.68

POSITIVE LOGITS

was

1.12

 weren

1.05

 wasn

1.04

 didn

1.02

Was

0.97

Wasn

0.95

 were

0.93

didn

0.92

had

0.91

Was

0.90

Activations Density 0.081%

negations and expressions of uncertainty or denial

No Comments

No Known Activations