INDEX

Explanations

specific terms related to certificates and documentation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 nakalista

-0.42

 включая

-0.32

 بما

-0.24

Personendaten

-0.24

 esboço

-0.24

ленно

-0.23

ubehör

-0.23

➌

-0.23

greenrobot

-0.23

 incluyendo

-0.22

POSITIVE LOGITS

 pleaſure

0.65

 juſ

0.64

 ſtre

0.61

 ProtoMessage

0.60

 occurs

0.60

iſchen

0.60

CloseOperation

0.60

Occurs

0.59

 leſs

0.59

 ddelwed

0.59

Activations Density 0.084%

specific terms related to certificates and documentation

No Comments

No Known Activations