INDEX

Explanations

expressions of admiration or positive descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

/Internal

-0.17

adesh

-0.17

izzle

-0.16

ilon

-0.16

 ÑĥÐ´Ð°

-0.15

alnÄ±z

-0.15

ekim

-0.15

 suce

-0.15

LayoutPanel

-0.14

Ð°Ð¹Ð´

-0.14

POSITIVE LOGITS

 mind

0.32

 simply

0.29

 beyond

0.29

jaw

0.28

jaw

0.27

 phen

0.26

 breath

0.25

 spell

0.25

 unlike

0.25

 Mind

0.25

Activations Density 0.242%

expressions of admiration or positive descriptions

No Comments

No Known Activations