INDEX

Explanations

terms that indicate measurement or evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ãĥ¼ãĥĳãĥ¼

-0.15

ç§°

-0.15

omu

-0.15

spar

-0.14

Collections

-0.14

.cgi

-0.14

ipse

-0.14

sty

-0.14

 Sabbath

-0.14

 Sabb

-0.13

POSITIVE LOGITS

.tm

0.16

OKIE

0.16

utex

0.15

assen

0.15

nett

0.15

ÏģÎ¹Î¬

0.14

ickle

0.14

_anchor

0.14

å¤§åħ¨

0.14

leccion

0.13

Activations Density 0.048%

terms that indicate measurement or evaluation

No Comments

No Known Activations