INDEX

Explanations

phrases related to discussions and specific topics

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

ReusableCell

-0.70

ictus

-0.50

ltä

-0.50

 peace

-0.46

 Fleming

-0.46

 part

-0.46

 cắt

-0.45

łych

-0.45

Pfe

-0.45

ایان

-0.45

POSITIVE LOGITS

 nakalista

1.00

 propOrder

0.97

InjectAttribute

0.92

 Bourgoin

0.91

MessageTagHelper

0.91

how

0.85

subpackage

0.83

 []).

0.82

setVerticalGroup

0.81

 متعلقه

0.81

Activations Density 0.198%

phrases related to discussions and specific topics

No Comments

No Known Activations