INDEX

Explanations

phrases related to resources, evidence, and support in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ÑĢÑĥÑĤ

-0.16

.nd

-0.15

å¹

-0.14

fly

-0.14

acock

-0.14

mans

-0.14

_frm

-0.14

Ð²ÑĸÑĤ

-0.14

_buckets

-0.13

jte

-0.13

POSITIVE LOGITS

uard

0.20

inski

0.15

 Deutsch

0.14

ousse

0.14

heits

0.14

rello

0.14

tom

0.14

ÑĨÐµÑĢ

0.14

ateg

0.14

allel

0.13

Activations Density 0.238%

phrases related to resources, evidence, and support in discussions

No Comments

No Known Activations