INDEX

Explanations

phrases related to news headlines or current events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "the"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

thood

-0.74

leeve

-0.64

iffe

-0.63

 Rahul

-0.60

 assume

-0.59

 Yuri

-0.58

 Gerard

-0.58

aba

-0.57

ALT

-0.57

 suppose

-0.57

POSITIVE LOGITS

ses

1.25

 same

1.16

 longest

1.06

 entire

1.05

 fastest

1.04

 slightest

1.03

 latter

1.03

 smallest

1.01

 hardest

0.97

 widest

0.97

Activations Density 0.332%

phrases related to news headlines or current events

occurrences of the word "the"

No Comments

No Known Activations