INDEX

Explanations

computer code related to manipulating data structures and algorithms

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ń·

-0.94

İĭ

-0.80

±

-0.77

Ļ½

-0.73

 obser

-0.70

 destro

-0.67

¿½

-0.67

acies

-0.66

onite

-0.66

©¶æ¥µ

-0.66

POSITIVE LOGITS

semble

0.76

REDACTED

0.76

LOG

0.73

byte

0.69

track

0.68

::

0.67

+)

0.67

*)

0.66

ept

0.64

perse

0.63

Activations Density 0.050%

computer code related to manipulating data structures and algorithms

No Comments

No Known Activations