INDEX

Explanations

numerical values and their associations in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

OrNull

-0.17

asher

-0.15

utton

-0.15

kit

-0.15

 paralysis

-0.14

ep

-0.14

contra

-0.14

 lifetime

-0.14

assa

-0.14

umps

-0.14

POSITIVE LOGITS

elsing

0.16

 Huff

0.15

.stamp

0.15

uces

0.14

-ret

0.14

strup

0.14

GINE

0.14

CRET

0.14

bane

0.14

COPE

0.14

Activations Density 0.004%

numerical values and their associations in the text

No Comments

No Known Activations