INDEX

Explanations

phrases related to learning and gaining knowledge

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

govtrack

-0.71

lished

-0.67

ordered

-0.64

orescent

-0.62

pei

-0.62

 shipments

-0.61

vati

-0.61

head

-0.60

conn

-0.60

heads

-0.59

POSITIVE LOGITS

 lessons

1.05

 lesson

1.04

learn

0.99

 learnt

0.92

Lear

0.92

 Teach

0.90

 Learned

0.88

 humility

0.85

 firsthand

0.83

 Curve

0.82

Activations Density 0.550%

phrases related to learning and gaining knowledge

No Comments

No Known Activations