INDEX

Explanations

terms related to HTTP requests and web programming

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

http requests and responses

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

Not in Any Lists

No Comments

Negative Logits

⟬

-0.64

IUrlHelper

-0.60

[@BOS@]

-0.59

<unused3>

-0.59

<unused23>

-0.59

<unused28>

-0.59

<unused17>

-0.59

<unused16>

-0.59

<unused8>

-0.59

<unused14>

-0.59

POSITIVE LOGITS

createServer

0.72

writeHead

0.49

 smtplib

0.42

http

0.41

HTTPS

0.35

 Server

0.34

devServer

0.34

Server

0.34

 Expédié

0.34

sidemargin

0.33

Activations Density 0.001%

terms related to HTTP requests and web programming

http requests and responses

No Comments

No Known Activations