INDEX

Explanations

items related to technical components and structures within programming code

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

colo

-0.14

dad

-0.14

reative

-0.14

hled

-0.14

Bounding

-0.14

/her

-0.13

istr

-0.13

igest

-0.13

daq

-0.13

baugh

-0.13

POSITIVE LOGITS

enticator

0.16

 Elect

0.15

Compat

0.15

untime

0.15

emsp

0.14

 mans

0.14

0.13

Exp

0.13

Exp

0.13

.prototype

0.13

Activations Density 0.117%

items related to technical components and structures within programming code

No Comments

No Known Activations