INDEX

Explanations

repeated mentions of the word "block" in various contexts

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

block

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

tableFuture

-0.65

 getItemId

-0.61

TokenNameDOT

-0.59

windowFixed

-0.57

faring

-0.57

prud

-0.55

genie

-0.53

 OnInit

-0.52

omitempty

-0.52

 GenerationType

-0.51

POSITIVE LOGITS

 block

2.17

 Block

1.97

block

1.93

Block

1.88

 BLOCK

1.79

 blocks

1.67

BLOCK

1.64

 Blocks

1.46

blocks

1.43

 blok

1.43

Activations Density 0.015%

repeated mentions of the word "block" in various contexts

block

No Comments

No Known Activations