INDEX

Explanations

phrases indicating singularity or uniqueness

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

raud

-0.16

lica

-0.16

cape

-0.16

aceous

-0.15

roje

-0.15

ible

-0.15

lickr

-0.14

anter

-0.14

ress

-0.14

igion

-0.14

POSITIVE LOGITS

ots

0.15

agers

0.15

-uri

0.15

yi

0.14

-third

0.14

áº£i

0.14

idas

0.14

ida

0.14

iloc

0.14

iner

0.14

Activations Density 0.059%

phrases indicating singularity or uniqueness

No Known Activations