INDEX

Explanations

names of locations and references to specific events or dates

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

agus

-0.70

ASE

-0.67

Â£

-0.64

endi

-0.64

pering

-0.61

winter

-0.61

ishers

-0.61

UGE

-0.61

liest

-0.60

adden

-0.58

POSITIVE LOGITS

 includes

0.87

 included

0.84

 referred

0.82

 served

0.80

 noted

0.78

 noteworthy

0.78

 risked

0.76

 known

0.75

 incorporates

0.74

 contributed

0.74

Activations Density 0.513%

names of locations and references to specific events or dates

No Comments

No Known Activations