INDEX

Explanations

mentions of tie-in products or collaborations

oai_token-act-pair · gpt-3.5-turbo

references to ties in various contexts

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

artney

-0.74

sburg

-0.73

ocate

-0.73

afa

-0.73

angel

-0.73

hma

-0.72

okin

-0.72

ocal

-0.70

azard

-0.69

olute

-0.69

POSITIVE LOGITS

tie

1.14

breaker

1.13

tie

1.13

 knot

1.02

 breaker

0.96

pipe

0.91

 tying

0.89

 tied

0.89

breakers

0.89

dye

0.87

Activations Density 0.013%

mentions of tie-in products or collaborations

references to ties in various contexts

No Comments

No Known Activations