INDEX

Explanations

If-then conditional statements

oai_token-act-pair · gpt-3.5-turbo

statements or conditions involving the verb "to be" in various forms, often indicating existence or state of being

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

avascript

-0.77

pedia

-0.73

Fish

-0.68

inav

-0.65

 Suite

-0.65

Fill

-0.64

çīĪ

-0.61

icular

-0.61

\\\\\\\\

-0.61

 ILCS

-0.60

POSITIVE LOGITS

 successful

1.13

 indeed

1.11

 ever

1.08

 able

1.00

 willing

0.97

 unsuccessful

0.97

 unable

0.95

 chosen

0.92

 allowed

0.92

 sufficiently

0.91

Activations Density 0.142%

If-then conditional statements

statements or conditions involving the verb "to be" in various forms, often indicating existence or state of being

No Comments

No Known Activations