INDEX

Explanations

references to social justice and accountability issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

kÃ¼

-0.15

 Ð¾Ð±ÑıÐ·

-0.14

pron

-0.14

ayo

-0.14

akis

-0.14

ÙģØ¶ÙĦ

-0.14

ÙĦØ§ÙĨ

-0.14

dur

-0.14

-0.13

*(*

-0.13

POSITIVE LOGITS

 harmless

0.42

 innoc

0.37

 innocent

0.36

 merely

0.30

 benign

0.29

 nothing

0.29

 legitimate

0.29

 innocence

0.28

 Innoc

0.27

 perfectly

0.27

Activations Density 0.436%

references to social justice and accountability issues

No Comments

No Known Activations