INDEX

Explanations

phrases related to options or alternatives

oai_token-act-pair · gpt-3.5-turbo

phrases that denote directions or methods of approach

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

usters

-0.78

ĸļ

-0.78

ulner

-0.75

mare

-0.67

Synopsis

-0.66

asts

-0.65

uster

-0.63

 Tanz

-0.61

mind

-0.61

ishable

-0.60

POSITIVE LOGITS

finding

0.84

point

0.83

 doors

0.68

sey

0.66

finder

0.66

seeing

0.65

 Houses

0.64

esa

0.64

ward

0.63

door

0.60

Activations Density 0.029%

phrases related to options or alternatives

phrases that denote directions or methods of approach

No Comments

No Known Activations