INDEX

Explanations

the occurrence of the prefix "Sh" in various contexts

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

ield

-0.19

jas

-0.19

 Luther

-0.16

#:

-0.15

shift

-0.15

volt

-0.15

ecc

-0.15

ort

-0.15

 blind

-0.15

 Shift

-0.15

POSITIVE LOGITS

oa

0.19

onda

0.18

Boom

0.15

kre

0.15

ermann

0.15

ady

0.15

cher

0.15

optimize

0.15

alom

0.15

oppers

0.15

Activations Density 0.027%

the occurrence of the prefix "Sh" in various contexts

No Comments

No Known Activations