INDEX

Explanations

locations or places named "Point."

oai_token-act-pair · gpt-3.5-turbo

references to specific geographical locations or points of interest

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

 delinqu

-0.85

natureconservancy

-0.78

ESE

-0.74

uthor

-0.74

 notor

-0.73

 destro

-0.73

Interstitial

-0.73

ÄŁ

-0.69

versely

-0.69

eco

-0.69

POSITIVE LOGITS

point

1.05

 Point

1.04

Point

0.86

hole

0.82

points

0.79

patrick

0.78

itude

0.77

 Bree

0.77

eous

0.74

 Forward

0.73

Activations Density 0.025%

locations or places named "Point."

references to specific geographical locations or points of interest

No Comments

No Known Activations