INDEX

Explanations

phrases related to events or actions that precede other events

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

 estekak

-0.82

 Thierry

-0.77

ništvo

-0.74

 Marl

-0.72

ראלי

-0.71

isOk

-0.70

 pido

-0.68

 Laramie

-0.68

ítmény

-0.67

 Volker

-0.66

POSITIVE LOGITS

 followed

0.82

CEED

0.81

 proceeding

0.81

companied

0.81

 accompanied

0.81

 PROCEEDINGS

0.77

 ressemb

0.76

 précé

0.74

following

0.73

》

0.72

Activations Density 0.223%

phrases related to events or actions that precede other events

No Comments

No Known Activations