INDEX

Explanations

words related to objects or individuals that are airborne or buoyant

oai_token-act-pair · gpt-3.5-turbo

instances of the word "floating" and related variations

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

heimer

-0.83

edy

-0.82

mins

-0.78

ifi

-0.75

anasia

-0.75

aldi

-0.74

ni

-0.74

uckle

-0.74

sen

-0.73

si

-0.72

POSITIVE LOGITS

 floating

0.99

 Floating

0.94

 floats

0.87

 afloat

0.76

 pond

0.76

 float

0.74

flo

0.74

Lag

0.74

 floated

0.73

 raft

0.72

Activations Density 0.011%

words related to objects or individuals that are airborne or buoyant

instances of the word "floating" and related variations

No Comments

No Known Activations