INDEX

Explanations

names and terms related to characters or titles in entertainment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

ghan

-0.15

erate

-0.14

ä»£

-0.14

efa

-0.14

 Mile

-0.13

colo

-0.13

DAQ

-0.13

 McMahon

-0.13

 Fuse

-0.13

 Laur

-0.12

POSITIVE LOGITS

'gc

0.18

owan

0.15

leigh

0.14

#af

0.14

pis

0.14

fal

0.14

@register

0.13

ffield

0.13

"urls

0.13

æ·»

0.13

Activations Density 0.043%

names and terms related to characters or titles in entertainment

No Comments

No Known Activations