INDEX

Explanations

phrases indicating minimal quantities or insignificance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

apesh

-0.19

nist

-0.16

 somewhere

-0.15

á»ĳt

-0.14

quina

-0.13

dun

-0.13

rej

-0.13

sert

-0.13

enet

-0.13

duto

-0.13

POSITIVE LOGITS

tons

0.19

/no

0.18

/small

0.18

/tiny

0.15

ening

0.15

ton

0.15

ened

0.14

TON

0.14

ort

0.14

ensor

0.14

Activations Density 0.022%

phrases indicating minimal quantities or insignificance

No Comments

No Known Activations