INDEX

Explanations

tokens related to the concept of ignoring or forgetting information

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

"ignore" or "forget"

np_acts-logits-general · gemini-2.0-flash

ignore version

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

WritableDatabase

-0.59

dotenv

-0.57

 oper

-0.55

Autowired

-0.53

-0.51

estad

-0.50

sa

-0.50

PathVariable

-0.50

 fær

-0.50

omaterial

-0.49

POSITIVE LOGITS

 Neglect

0.85

 виправивши

0.83

 ignoring

0.80

 FORGET

0.79

 Ignore

0.77

 ignores

0.76

 forgetting

0.75

AddTagHelper

0.75

 Forget

0.75

掉

0.74

Activations Density 0.208%

tokens related to the concept of ignoring or forgetting information

"ignore" or "forget"

ignore version

No Comments

No Known Activations