INDEX

Explanations

details related to individuals, organizations, and events within a specific context or narrative

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

utations

-0.70

ezvous

-0.68

resy

-0.63

oxicity

-0.63

 wonders

-0.62

bernatorial

-0.61

gment

-0.60

leground

-0.60

:=

-0.60

 silence

-0.59

POSITIVE LOGITS

 onwards

0.81

 Architects

0.81

iets

0.72

who

0.70

 beforehand

0.69

ÃŃs

0.66

0.64

orks

0.64

 during

0.64

 consortium

0.64

Activations Density 2.721%

details related to individuals, organizations, and events within a specific context or narrative

No Comments

No Known Activations