INDEX

Explanations

occurrences of events, presentations, and activities in a structured format

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iteral

-0.13

Ã¡Äį

-0.13

ycz

-0.12

IgnoreCase

-0.12

à¸²à¸

-0.12

Ã¼st

-0.12

 Äĳá»Ŀi

-0.12

ÅĤaw

-0.12

 kategor

-0.12

ãĥĥãĥģ

-0.11

POSITIVE LOGITS

 ending

0.77

 finish

0.76

 conclude

0.73

 conclusion

0.73

 finishes

0.72

 concludes

0.71

 finishing

0.71

 ends

0.71

 concluded

0.68

 ended

0.67

Activations Density 0.918%

occurrences of events, presentations, and activities in a structured format

No Comments

No Known Activations