INDEX

Explanations

terms related to user-friendliness and supportive systems

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

argins

-0.15

 strt

-0.15

VisualStyle

-0.14

 refin

-0.14

 Cort

-0.14

ustos

-0.13

 Cout

-0.13

iena

-0.13

¬Ĥ

-0.13

quiv

-0.13

POSITIVE LOGITS

èĮĥ

0.16

ÑģÑĤÐ¾ÑĢ

0.15

/octet

0.15

ÐµÑĢÐµÐ¶

0.14

æ¹

0.13

ennen

0.13

nest

0.13

entin

0.13

MOTE

0.13

oron

0.13

Activations Density 0.102%

terms related to user-friendliness and supportive systems

No Comments

No Known Activations