INDEX

Explanations

specific names and identifiers related to food items or biological compounds

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

 CreateTagHelper

-0.78

UnusedPrivate

-0.77

 &___

-0.73

 AssemblyTitle

-0.73

verifyException

-0.71

complexContent

-0.69

цездатний

-0.68

 unknownFields

-0.67

 informée

-0.67

GOTREF

-0.66

POSITIVE LOGITS

is

0.42

sat

0.40

 will

0.36

 actually

0.35

 mukaan

0.35

 adalah

0.35

 provides

0.35

这句话

0.34

 establishes

0.34

are

0.34

Activations Density 0.033%

specific names and identifiers related to food items or biological compounds

No Comments

No Known Activations