INDEX

Explanations

numbers and their contexts in various formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

UnknownFields

-0.86

../../../

-0.84

下载附件

-0.83

ərbaycan

-0.82

 Walkover

-0.82

 للمعارف

-0.76

 szóci

-0.76

']))

-0.73

ので

-0.73

omiast

-0.71

POSITIVE LOGITS

rd

1.73

1.11

GPP

0.85

rd

0.83

ßig

0.80

rds

0.77

0.76

prnewswire

0.76

RD

0.74

Activations Density 1.295%

numbers and their contexts in various formats

No Comments

No Known Activations