INDEX

Explanations

coding-related terms in configuration and programming contexts

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

 Mant

-0.16

-0.15

entifier

-0.15

bef

-0.15

datap

-0.15

ry

-0.14

ari

-0.14

uluk

-0.14

 Dans

-0.14

ÃŃnu

-0.14

POSITIVE LOGITS

asher

0.16

chio

0.15

itten

0.15

Constants

0.14

.TypeOf

0.14

ete

0.14

asel

0.14

chop

0.14

.YesNo

0.14

constants

0.13

Activations Density 0.054%

coding-related terms in configuration and programming contexts

No Comments

No Known Activations