INDEX

Explanations

references to inns and motels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

inea

-0.15

acho

-0.15

conj

-0.14

erk

-0.14

rok

-0.13

/wait

-0.13

aft

-0.13

alo

-0.13

achs

-0.13

ona

-0.13

POSITIVE LOGITS

earer

0.16

Spi

0.15

idal

0.15

ri

0.14

rico

0.14

TintColor

0.14

-svg

0.14

rising

0.14

ambi

0.14

sco

0.13

Activations Density 0.007%

references to inns and motels

No Comments

No Known Activations