INDEX

Explanations

phrases indicating estimates or approximations of quantity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

uled

-0.17

istani

-0.16

agara

-0.15

lÃ¡

-0.15

/stretch

-0.15

etc

-0.13

lein

-0.13

tha

-0.13

.bb

-0.13

ses

-0.13

POSITIVE LOGITS

lier

0.17

adel

0.15

inis

0.15

LY

0.14

akk

0.14

;y

0.14

eo

0.14

EO

0.14

 mesmo

0.14

Activations Density 0.084%

phrases indicating estimates or approximations of quantity

No Comments

No Known Activations