INDEX

Explanations

keywords related to the phrase "works for" in various contexts

oai_token-act-pair · gpt-3.5-turbo

the occurrence of the word "works" 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

odder

-0.67

Cath

-0.66

SPONSORED

-0.66

cele

-0.66

Lay

-0.65

antha

-0.64

angular

-0.64

è£ħ

-0.63

Ce

-0.63

arta

-0.61

POSITIVE LOGITS

hops

1.44

paces

1.32

heet

1.29

hirt

1.10

hift

0.96

pace

0.92

bench

0.87

omething

0.82

ivism

0.81

manship

0.80

Activations Density 0.034%

keywords related to the phrase "works for" in various contexts

the occurrence of the word "works" in various contexts

No Comments

No Known Activations