INDEX

Explanations

themes related to judgment and accountability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gregated

-0.17

 Nacht

-0.17

edral

-0.16

upal

-0.16

esini

-0.15

exo

-0.15

azio

-0.15

iza

-0.15

itaire

-0.15

alama

-0.15

POSITIVE LOGITS

 harsh

0.16

 charged

0.15

mention

0.15

CTS

0.15

ä¹³

0.15

 bean

0.14

TS

0.14

 Middle

0.14

avel

0.14

ng

0.13

Activations Density 0.246%

themes related to judgment and accountability

No Comments

No Known Activations