INDEX

Explanations

measurements of weight and dimensions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

bis

-0.16

rost

-0.16

-0.15

idar

-0.15

 birthday

-0.15

irs

-0.14

 cold

-0.14

Ins

-0.14

ij

-0.14

 Grill

-0.14

POSITIVE LOGITS

@student

0.16

mdi

0.16

ÎłÎĳ

0.15

ltre

0.14

ewood

0.14

ourd

0.14

rrha

0.14

acci

0.14

ÐµÑĩÐ½Ð¾

0.13

ITTE

0.13

Activations Density 0.008%

measurements of weight and dimensions

No Comments

No Known Activations