INDEX

Explanations

references to specific years, particularly those in the late 1990s

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

alus

-0.17

alis

-0.16

lection

-0.16

CADE

-0.16

presso

-0.15

asl

-0.15

ifton

-0.15

enc

-0.15

enc

-0.15

POSITIVE LOGITS

ï¼Īå¹³æĪĲ

0.17

INGS

0.17

à¤®à¤¤

0.15

éº¼

0.15

stown

0.15

kus

0.15

Ø±ÛĮØ¨

0.14

ourd

0.14

INGLE

0.14

pedia

0.14

Activations Density 0.033%

references to specific years, particularly those in the late 1990s

No Comments

No Known Activations