INDEX

Explanations

phrases indicating the best or most optimal course of action or solution

oai_token-act-pair · gpt-3.5-turbo

phrases that assert superiority or optimality in various contexts

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

ancies

-0.76

each

-0.73

thood

-0.70

eds

-0.69

 teasp

-0.67

sleep

-0.66

oops

-0.64

events

-0.63

scl

-0.63

arms

-0.62

POSITIVE LOGITS

 thing

1.14

 antidote

1.03

 solution

1.02

 indication

1.00

 indicator

0.99

 source

0.98

 reason

0.97

 option

0.96

 predictor

0.94

 motiv

0.92

Activations Density 0.178%

phrases indicating the best or most optimal course of action or solution

phrases that assert superiority or optimality in various contexts

No Comments

No Known Activations