INDEX

Explanations

geographical or place-based references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

oad

-0.16

POSE

-0.16

artz

-0.15

ulis

-0.15

brief

-0.15

intree

-0.15

apesh

-0.15

BuilderInterface

-0.15

uzzi

-0.14

rapper

-0.14

POSITIVE LOGITS

tal

0.22

 tang

0.21

Tal

0.21

tal

0.20

wald

0.17

 kilomet

0.16

 berg

0.16

 Hang

0.15

ent

0.15

esco

0.15

Activations Density 0.047%

geographical or place-based references

No Comments

No Known Activations

geographical or place-based references

No Comments

No Known Activations