INDEX

Explanations

words related to symbols in code languages like C and C++

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ORED

-1.01

 guiActiveUnfocused

-0.90

ãģ®éŃĶ

-0.76

ishly

-0.69

 Dinosaur

-0.65

 actionGroup

-0.64

 admission

-0.63

-0.61

sburgh

-0.61

 resort

-0.60

POSITIVE LOGITS

bil

1.26

ringe

1.20

mbol

1.17

nerg

1.08

thia

1.07

ntax

1.07

philis

1.05

pher

1.05

rien

0.98

rians

0.97

Activations Density 1.008%

words related to symbols in code languages like C and C++

No Comments

No Known Activations