INDEX

Explanations

technical markup or coding syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

legg

-0.16

_sqrt

-0.15

 intermedi

-0.14

 bush

-0.14

çĬ¶

-0.14

GuidId

-0.14

byname

-0.13

 Episodes

-0.13

 intens

-0.13

strftime

-0.13

POSITIVE LOGITS

 offset

0.31

 offsets

0.28

-offset

0.27

Offset

0.27

 Offset

0.26

offset

0.25

pos

0.25

pos

0.25

_offset

0.25

Pos

0.25

Activations Density 0.466%

technical markup or coding syntax

No Comments

No Known Activations