INDEX

Explanations

specific adjectives that describe actions or characteristics, such as "sparing", "terrifying", "vicious", "exact", and "leisure"

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

anders

-0.79

APH

-0.66

axy

-0.66

REL

-0.65

DonaldTrump

-0.64

ploma

-0.63

aden

-0.61

ARM

-0.61

ODUCT

-0.61

iltr

-0.60

POSITIVE LOGITS

ly

2.93

LY

1.84

lys

1.42

edly

1.31

liness

1.31

lies

1.25

fully

1.21

ity

1.16

ously

1.15

lly

1.15

Activations Density 1.945%

specific adjectives that describe actions or characteristics, such as "sparing", "terrifying", "vicious", "exact", and "leisure"

No Comments

No Known Activations