INDEX

Explanations

quantities and numerical references

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

Not in Any Lists

No Comments

Negative Logits

ourn

-0.19

fleet

-0.15

æľĭ

-0.15

acionales

-0.15

ichtet

-0.14

hoot

-0.14

leo

-0.14

rud

-0.14

ardin

-0.13

/helpers

-0.13

POSITIVE LOGITS

èª

0.16

 Chambers

0.14

fold

0.14

eyh

0.14

awai

0.13

GBK

0.13

mart

0.13

uerdo

0.13

 getChild

0.13

roke

0.13

Activations Density 0.117%

quantities and numerical references

No Comments

No Known Activations