INDEX

Explanations

mathematical expressions and operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

hod

-0.16

_CONTINUE

-0.15

å§ĵ

-0.14

anj

-0.14

ijing

-0.13

ouch

-0.13

ana

-0.13

otte

-0.13

Ao

-0.13

imp

-0.13

POSITIVE LOGITS

insky

0.16

pollo

0.16

skirts

0.15

inesis

0.15

å·§

0.15

Kub

0.14

bane

0.14

ucks

0.14

ADX

0.14

rrha

0.13

Activations Density 0.071%

mathematical expressions and operations

No Comments

No Known Activations