INDEX

Explanations

references to qualifications, titles, or roles of individuals in a professional context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ibel

-0.15

 bÄĥng

-0.15

 footage

-0.15

 flesh

-0.14

stamp

-0.14

ÑĮÑİÑĤ

-0.14

 Stamp

-0.14

 Gang

-0.13

iod

-0.13

oni

-0.13

POSITIVE LOGITS

stile

0.15

Ø¬Ø©

0.15

$MESS

0.14

iners

0.14

icus

0.14

rias

0.14

enderror

0.14

isa

0.14

 AUTHORS

0.14

ãĤ«ãĥĨ

0.14

Activations Density 0.016%

references to qualifications, titles, or roles of individuals in a professional context

No Comments

No Known Activations