INDEX

Explanations

sections of text that include menu and navigation elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 bows

-0.17

áº¯t

-0.15

.plus

-0.15

aring

-0.15

ima

-0.15

iew

-0.15

bow

-0.15

bow

-0.14

eur

-0.14

eka

-0.14

POSITIVE LOGITS

lak

0.15

laz

0.15

 Shank

0.15

anela

0.14

_wallet

0.14

jac

0.14

tÃŃ

0.14

multiline

0.14

ovÃŃ

0.13

ONSE

0.13

Activations Density 0.011%

sections of text that include menu and navigation elements

No Comments

No Known Activations