INDEX

Explanations

references to support and recognition associated with events or initiatives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ohl

-0.19

ÏĩÎ®

-0.16

iaz

-0.15

.mods

-0.15

opa

-0.14

Î»ÎŃÎ¿Î½

-0.14

 engulf

-0.14

.Typed

-0.13

agli

-0.13

adden

-0.13

POSITIVE LOGITS

pch

0.16

ibus

0.15

Ni

0.15

yo

0.15

ulo

0.15

ottom

0.15

Ni

0.14

Duy

0.14

etwork

0.14

Err

0.14

Activations Density 0.089%

references to support and recognition associated with events or initiatives

No Comments

No Known Activations