INDEX

Explanations

punctuation and specific phrases or segments that indicate decision-making or significant moments of change

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 represented

-0.19

 meant

-0.18

represented

-0.16

 portrayed

-0.16

 depicted

-0.15

 preceded

-0.15

 seen

-0.14

 presented

-0.14

 supported

-0.14

 Seen

-0.14

POSITIVE LOGITS

ate

0.29

 took

0.26

 went

0.26

 drank

0.26

 threw

0.25

got

0.25

 blew

0.24

went

0.24

 flew

0.24

 withdrew

0.24

Activations Density 0.201%

punctuation and specific phrases or segments that indicate decision-making or significant moments of change

No Comments

No Known Activations