INDEX

Explanations

references to educational or professional contexts, particularly involving interactions and evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

areth

-0.18

Ð¼ÐµÐ½Ð°

-0.15

Dit

-0.15

Î±ÏģÎ¬

-0.14

atha

-0.14

obot

-0.14

ãĤ·ãĥ§

-0.14

icken

-0.14

menin

-0.14

swire

-0.14

POSITIVE LOGITS

 therein

0.28

åħ¶ä¸Ń

0.27

 thereof

0.21

ãģĿãģĵ

0.21

è¯¥

0.20

 dort

0.20

 dess

0.20

 ÑĤÐ°Ð¼

0.19

éĤ£éĩĮ

0.19

 daar

0.19

Activations Density 0.438%

references to educational or professional contexts, particularly involving interactions and evaluations

No Comments

No Known Activations