INDEX

Explanations

phrases indicating connections or relationships among elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

stantiate

-0.15

ffb

-0.15

/or

-0.14

ses

-0.14

uars

-0.14

enne

-0.13

ÏĦÎµÏį

-0.13

/her

-0.13

shi

-0.13

 Aires

-0.13

POSITIVE LOGITS

orem

0.18

oretical

0.16

amp

0.16

ermal

0.15

iko

0.15

visa

0.14

ories

0.14

odor

0.14

ENUM

0.14

Activations Density 0.248%

phrases indicating connections or relationships among elements

No Comments

No Known Activations