INDEX

Explanations

blocks of code or structured text elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

osemite

-0.15

 Profession

-0.14

ToDevice

-0.14

 Crane

-0.14

iti

-0.14

 Profes

-0.13

iÅ¡tÄĽ

-0.13

å¾®è½¯éĽħé»ĳ

-0.13

.au

-0.13

 crossorigin

-0.13

POSITIVE LOGITS

ecz

0.15

Å¡Å¥

0.15

ey

0.15

OSH

0.14

Disclosure

0.14

ammer

0.14

bey

0.14

ress

0.14

Til

0.13

Activations Density 0.004%

blocks of code or structured text elements

No Comments

No Known Activations