INDEX

Explanations

specific numerical or formatted values, particularly related to data structures or programming constructs

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

oday

-0.16

Twe

-0.14

 Tham

-0.14

à¥įà¤ł

-0.14

 cháº£y

-0.13

ProgressHUD

-0.13

Tat

-0.13

tbl

-0.13

 Tahoe

-0.13

 Tanner

-0.13

POSITIVE LOGITS

0.35

 transfer

0.20

Û±Û°

0.19

 Transfer

0.17

transfer

0.17

ayah

0.17

Ð¾ÑĤ

0.17

-transfer

0.16

Transfer

0.16

ot

0.16

Activations Density 0.031%

specific numerical or formatted values, particularly related to data structures or programming constructs

No Comments

No Known Activations