INDEX

Explanations

contextual references to the term "this."

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

this followed by code or punctuation

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

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

IFrame

Link

Negative Logits

 ſei

-0.79

<unused14>

-0.78

[@BOS@]

-0.78

<unused8>

-0.78

<unused68>

-0.78

<unused79>

-0.78

<unused43>

-0.78

<unused42>

-0.78

<unused23>

-0.78

<unused41>

-0.78

POSITIVE LOGITS

this

0.87

This

0.65

THIS

0.61

 this

0.60

 This

0.59

 THIS

0.58

 acest

0.48

self

0.45

 diesem

0.45

 acestui

0.45

Activations Density 0.006%

contextual references to the term "this."

this followed by code or punctuation

No Known Activations

contextual references to the term "this."

this followed by code or punctuation

No Known Activations