INDEX

Explanations

valid hexadecimal codes

oai_token-act-pair · gpt-3.5-turbo

occurrences of a specific character or symbol

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wagen

-0.87

entious

-0.77

ritic

-0.76

aceutical

-0.75

izabeth

-0.75

heid

-0.75

idy

-0.74

mie

-0.72

imony

-0.72

ardless

-0.72

POSITIVE LOGITS

LAB

0.85

 magnification

0.78

 Expand

0.76

ghai

0.66

Python

0.66

Brew

0.64

[[

0.62

Ģ

0.62

_>

0.61

Pict

0.60

Activations Density 0.016%

valid hexadecimal codes

occurrences of a specific character or symbol

No Comments

No Known Activations