INDEX

Explanations

verbs ending in 'ape' with varying degrees of specificity

oai_token-act-pair · gpt-3.5-turbo

references to the word "ape" 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Beir

-0.86

quickShipAvailable

-0.76

sch

-0.75

naires

-0.75

naire

-0.74

ichick

-0.70

-0.69

SLI

-0.69

 McMaster

-0.67

SOURCE

-0.67

POSITIVE LOGITS

utic

1.16

utics

1.03

etus

1.00

llan

0.82

lla

0.79

apers

0.76

ople

0.75

rine

0.72

Ã±

0.72

ographed

0.72

Activations Density 0.013%

verbs ending in 'ape' with varying degrees of specificity

references to the word "ape" in various contexts

No Comments

No Known Activations