INDEX

Explanations

words related to large quantities or intensities

oai_token-act-pair · gpt-3.5-turbo

instances of the word "output" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tarian

-0.68

 Clause

-0.65

 Sharif

-0.65

roy

-0.64

è£ħ

-0.63

 viol

-0.62

tone

-0.62

stan

-0.62

 clauses

-0.60

 statically

-0.60

POSITIVE LOGITS

ouring

1.45

acing

1.11

oring

1.03

ours

1.02

ored

1.02

ifts

1.00

acement

0.99

acements

0.96

oured

0.96

orer

0.95

Activations Density 0.051%

words related to large quantities or intensities

instances of the word "output" and its variations

No Comments

No Known Activations