INDEX

Explanations

references to cycling or cycles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ÑĤÐ¸Ð½

-0.15

down

-0.15

ness

-0.15

Ã¸y

-0.14

ills

-0.14

iesta

-0.14

lao

-0.14

Sig

-0.14

Ill

-0.14

POSITIVE LOGITS

mpl

0.17

indrical

0.16

hood

0.15

hone

0.15

udad

0.15

.mount

0.15

uden

0.15

sik

0.14

licity

0.14

lops

0.14

Activations Density 0.048%

references to cycling or cycles

No Comments

No Known Activations