INDEX

Explanations

sequences of letters that match specific patterns or codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

-0.56

Di

-0.56

or

-0.54

-0.53

re

-0.52

-0.51

-0.50

-0.49

POSITIVE LOGITS

 myſelf

1.21

 leſs

1.16

 ་་

1.16

ſelf

1.13

 ſever

1.11

 unſ

1.10

 itſelf

1.10

ſelves

1.10

AddTagHelper

1.09

 pleaſure

1.08

Activations Density 0.021%

sequences of letters that match specific patterns or codes

No Comments

No Known Activations