INDEX

Explanations

phrases related to spending time

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing totality or completeness

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

aminer

-0.71

 Kamp

-0.69

nom

-0.68

IDS

-0.66

ufact

-0.65

licts

-0.65

grad

-0.65

 Peninsula

-0.62

lict

-0.62

lav

-0.61

POSITIVE LOGITS

ocating

1.13

udes

1.12

usions

1.09

igators

1.08

igator

1.08

 kinds

1.08

uding

1.04

 sorts

1.02

usion

0.94

ude

0.93

Activations Density 0.139%

phrases related to spending time

phrases emphasizing totality or completeness

No Comments

No Known Activations