INDEX

Explanations

words related to legal proceedings and activities

oai_token-act-pair · gpt-3.5-turbo

significant actions or occurrences within a narrative or commentary

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.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

venge

-0.66

 Reverse

-0.64

 accelerating

-0.62

coming

-0.60

 boosting

-0.58

 transforming

-0.56

 bullish

-0.55

 Gain

-0.55

bringing

-0.54

 acceleration

-0.54

POSITIVE LOGITS

 merely

1.35

 instead

1.06

Instead

1.05

 simply

1.03

 only

0.99

Nevertheless

0.95

only

0.91

 Instead

0.90

Nonetheless

0.88

opol

0.87

Activations Density 0.842%

words related to legal proceedings and activities

significant actions or occurrences within a narrative or commentary

No Comments

No Known Activations