INDEX

Explanations

phrases expressing suggestions or recommendations

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

ÑıÑĩ

-0.17

bidden

-0.16

ambi

-0.16

azeera

-0.15

PÅĻi

-0.15

inburgh

-0.15

.const

-0.15

_marshall

-0.15

altung

-0.15

dech

-0.15

POSITIVE LOGITS

tro

0.15

Tu

0.14

alse

0.14

coe

0.14

Vi

0.14

 Laud

0.14

airo

0.14

åĪ¥

0.14

ÙıØ¹

0.13

Pel

0.13

Activations Density 0.017%

phrases expressing suggestions or recommendations

No Comments

No Known Activations