INDEX

Explanations

expressions of personal experience and recovery

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Immediately

-0.16

 beforehand

-0.15

Immediately

-0.14

 preceding

-0.14

ande

-0.14

ita

-0.14

onian

-0.14

åķ

-0.14

osa

-0.14

Ð´Ñĭ

-0.14

POSITIVE LOGITS

 eventually

0.79

 Eventually

0.73

Eventually

0.71

 eventual

0.66

 soon

0.59

 Soon

0.49

 ultimately

0.48

 gradually

0.47

Soon

0.46

soon

0.43

Activations Density 0.563%

expressions of personal experience and recovery

No Comments

No Known Activations