INDEX

Explanations

preposition 'for' followed by a verb

oai_token-act-pair · gpt-3.5-turbo

phrases indicating causation or purpose

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

nikov

-0.73

buster

-0.70

Strange

-0.70

orbit

-0.65

omi

-0.64

Luc

-0.62

MJ

-0.62

marine

-0.62

illin

-0.61

olan

-0.60

POSITIVE LOGITS

 purposes

1.26

 example

1.16

 instance

1.14

cible

1.10

geries

1.10

bidden

1.09

 sake

1.05

 reasons

1.05

gotten

1.05

 centuries

1.02

Activations Density 0.310%

preposition 'for' followed by a verb

phrases indicating causation or purpose

No Comments

No Known Activations