INDEX

Explanations

phrases indicating presence or availability

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

chod

-0.15

ÐŁÐļ

-0.15

Ã¨

-0.15

hee

-0.15

sson

-0.14

 liner

-0.14

gee

-0.14

awns

-0.14

pton

-0.13

Eur

-0.13

POSITIVE LOGITS

0.15

istle

0.15

-sizing

0.14

 Nose

0.14

fang

0.14

grese

0.14

oops

0.14

ilded

0.14

okes

0.14

ISTR

0.13

Activations Density 0.019%

phrases indicating presence or availability

No Comments

No Known Activations