INDEX

Explanations

mathematical terminology related to equations and structures

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

ormsg

-0.18

chwitz

-0.16

tains

-0.15

azzi

-0.14

yon

-0.14

ãĥ£

-0.14

yre

-0.14

 OnTrigger

-0.14

odiac

-0.13

yms

-0.13

POSITIVE LOGITS

.borrow

0.15

Ear

0.14

Ear

0.14

èĢ³

0.13

_hdl

0.13

imple

0.13

 susp

0.13

uctose

0.13

ear

0.13

ank

0.13

Activations Density 0.022%

mathematical terminology related to equations and structures

No Comments

No Known Activations