INDEX

Explanations

structured lists or enumerations

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

-0.19

icorn

-0.16

.increment

-0.15

bons

-0.14

esting

-0.14

suf

-0.14

³

-0.14

apos

-0.14

iro

-0.13

ctl

-0.13

POSITIVE LOGITS

¯u

0.16

ackage

0.16

ospels

0.15

rowable

0.15

_PKG

0.14

SystemService

0.14

SocketAddress

0.14

Î¶Îµ

0.14

asca

0.14

 ÙħØ±Ø§Øª

0.14

Activations Density 0.007%

structured lists or enumerations

No Comments

No Known Activations