INDEX

Explanations

references to completing tasks or accomplishing work

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

çµ

-0.14

oire

-0.14

Dash

-0.14

ãģ¤

-0.13

Opened

-0.13

gri

-0.13

ãĥĥãĥĦ

-0.13

readcr

-0.13

ernaut

-0.13

fully

-0.13

POSITIVE LOGITS

 done

0.86

done

0.73

 Done

0.72

-done

0.68

Done

0.67

 DONE

0.63

.done

0.61

	done

0.60

_done

0.58

DONE

0.50

Activations Density 0.117%

references to completing tasks or accomplishing work

No Comments

No Known Activations