INDEX

Explanations

dates and related information

oai_token-act-pair · gpt-3.5-turbo

instances of individuals' names and relevance markers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ancial

-0.75

atile

-0.72

TEXTURE

-0.68

ettings

-0.68

itized

-0.67

adh

-0.65

alyses

-0.64

atic

-0.63

heastern

-0.63

etary

-0.62

POSITIVE LOGITS

who

1.26

 whose

1.25

who

1.12

whose

1.12

 nicknamed

1.11

 whom

1.05

 meanwhile

1.03

 born

0.97

 formerly

0.95

Activations Density 0.125%

dates and related information

instances of individuals' names and relevance markers

No Comments

No Known Activations