INDEX

Explanations

occurrences of numerical values or measurements in the text

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

Empty line or newline characters that serve as section separators in structured documents such as code files, academic papers, or technical documentation.

eleuther_acts_top20 · claude-3-7-sonnet-20250219 Triggered by @lu-christina

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

RenderAtEndOf

-1.42

iſen

-1.42

 queſta

-1.41

ſicht

-1.37

ロウィン

-1.34

<unused68>

-1.34

<unused14>

-1.34

<unused28>

-1.34

<unused8>

-1.34

[@BOS@]

-1.34

POSITIVE LOGITS

0.68

0.58

0.57

↵↵

0.57

The

0.56

0.51

0.50

0.48

0.47

Activations Density 0.001%

occurrences of numerical values or measurements in the text

Empty line or newline characters that serve as section separators in structured documents such as code files, academic papers, or technical documentation.

No Comments

No Known Activations