INDEX

Explanations

terms related to immune responses and associated biological processes

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

 يتيمه

-0.47

 Савезне

-0.43

EDEFAULT

-0.42

 queſta

-0.41

 sweeteners

-0.40

ikala

-0.40

enterOuterAlt

-0.40

complexContent

-0.40

 schoenen

-0.39

riwal

-0.39

POSITIVE LOGITS

 défense

0.66

 defence

0.59

 defense

0.58

 defensa

0.55

 defesa

0.49

 combatir

0.48

 combat

0.48

 patrolling

0.47

 defences

0.47

defense

0.47

Activations Density 0.161%

terms related to immune responses and associated biological processes

No Comments

No Known Activations