INDEX

Explanations

conjunctions connecting two related ideas

oai_token-act-pair · gpt-3.5-turbo

instances of the word "And."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

fell

-0.77

ãģĨ

-0.71

ses

-0.69

-0.65

DNA

-0.65

atical

-0.64

sk

-0.64

-0.62

arthed

-0.61

å£

-0.61

POSITIVE LOGITS

romeda

1.16

rea

1.12

hra

1.06

alus

1.01

 furthermore

1.00

 secondly

1.00

ersen

0.88

 moreover

0.87

 therein

0.80

 consequently

0.78

Activations Density 0.070%

conjunctions connecting two related ideas

instances of the word "And."

No Comments

No Known Activations