INDEX

Explanations

lines or blocks of code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits



-1.24

)";

-1.16

 />';

-1.10

$.

-1.07

 ―――――

-1.06

$_"

-1.06

"):

-1.05

 iſt

-1.05

 Jefus

-1.04

 ...]

-1.01

POSITIVE LOGITS

1.19

0.86

0.69

0.65

0.63

0.57

0.56

0.50

","

0.49

Activations Density 0.148%

lines or blocks of code

No Comments

No Known Activations