INDEX

Explanations

references to time periods and transitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 termin

-0.16

 Finish

-0.16

inish

-0.16

 finishes

-0.15

 earlier

-0.15

 finishing

-0.15

ear

-0.15

 Shannon

-0.14

io

-0.14

 Payne

-0.14

POSITIVE LOGITS

 next

0.35

next

0.31

 nÃ¤ch

0.25

-next

0.25

 NEXT

0.25

 nÃ¤chsten

0.24

(next

0.24

 Next

0.24

_next

0.24

 à¤ħà¤Ĺà¤²

0.23

Activations Density 0.148%

references to time periods and transitions

No Comments

No Known Activations