INDEX

Explanations

terms related to health conditions and their causal relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

-0.60

can

-0.52

 Anda

-0.50

 super

-0.50

 normal

-0.48

off

-0.47

dis

-0.47

é

-0.47

 inter

-0.47

don

-0.46

POSITIVE LOGITS

 Jefus

1.05

Efq

1.00

 itſelf

0.97

 Monfieur

0.93

 myſelf

0.90

 enfans

0.88

 ſeveral

0.87

 purpoſe

0.85

 Majefty

0.85

 themſelves

0.84

Activations Density 1.032%

terms related to health conditions and their causal relationships

No Comments

No Known Activations