INDEX

Explanations

prominent names or titles, particularly in the context of entertainment or cultural references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ryn

-0.19

irth

-0.16

 treff

-0.14

ãĥ¼ãĥł

-0.14

empor

-0.14

leitung

-0.14

ombok

-0.14

ursive

-0.13

insula

-0.13

Keeper

-0.13

POSITIVE LOGITS

Pat

0.21

pat

0.20

Pat

0.19

_pat

0.18

.pat

0.17

PAT

0.16

(pat

0.15

pat

0.15

Ø§ÙĤÙĦ

0.15

ItemSelectedListener

0.14

Activations Density 0.030%

prominent names or titles, particularly in the context of entertainment or cultural references

No Comments

No Known Activations