INDEX

Explanations

proper nouns, particularly names and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

istros

-0.16

uko

-0.14

prend

-0.14

utin

-0.13

stakes

-0.13

UBL

-0.13

chn

-0.13

intree

-0.13

onde

-0.13

Kab

-0.13

POSITIVE LOGITS

deen

0.15

ÙĬØ§Øª

0.14

referer

0.14

WEEN

0.14

ĭ

0.13

 apart

0.13

ittance

0.13

ussen

0.13

ceived

0.13

 replic

0.13

Activations Density 0.054%

proper nouns, particularly names and organizations

No Comments

No Known Activations