INDEX

Explanations

descriptive adjectives related to personalities and characteristics

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

ãĥĻ

-0.64

ragon

-0.64

quist

-0.63

uci

-0.63

OTHER

-0.62

Otherwise

-0.62

ew

-0.61

etime

-0.61

é¾į

-0.61

OND

-0.61

POSITIVE LOGITS

 however

0.79

huh

0.76

eh

0.72

 moreover

0.67

 these

0.63

 this

0.60

 Garland

0.60

 though

0.60

it

0.59

0.58

Activations Density 1.144%

descriptive adjectives related to personalities and characteristics

No Comments

No Known Activations