INDEX

Explanations

phrases related to physical locations or objects

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

ãĤ¹ãĥĪ

-0.75

ãĤ£

-0.67

ãĥī

-0.67

Deal

-0.61

Introduced

-0.60

è¦ļéĨĴ

-0.60

Access

-0.59

yah

-0.59

ãĥı

-0.58

POSITIVE LOGITS

ground

1.22

mentioned

0.86

grade

0.77

deck

0.74

 grade

0.73

 ground

0.73

neath

0.70

ebin

0.69

hand

0.69

perty

0.69

Activations Density 0.297%

phrases related to physical locations or objects

No Comments

No Known Activations