INDEX

Explanations

phrases related to potential future events or consequences

oai_token-act-pair · gpt-3.5-turbo

the word "will" in various contexts, indicating future events or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Sandwich

-0.67

quickShipAvailable

-0.65

soType

-0.64

 Waterloo

-0.61

 Laden

-0.60

 Modified

-0.60

 predecessor

-0.59

 srfAttach

-0.59

HQ

-0.58

 expires

-0.58

POSITIVE LOGITS

 flock

1.20

 suffer

1.15

 doubtless

1.12

 perceive

1.12

be

1.12

 attest

1.11

 gladly

1.11

 undoubtedly

1.06

 recognize

1.05

 realize

1.05

Activations Density 0.228%

phrases related to potential future events or consequences

the word "will" in various contexts, indicating future events or conditions

No Comments

No Known Activations