INDEX

Explanations

terms related to a certain level or limit (a "threshold")

oai_token-act-pair · gpt-3.5-turbo

references to quantitative limits or levels of measurement

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

ulf

-0.77

ocate

-0.75

ocation

-0.70

emis

-0.67

ONG

-0.67

ocated

-0.65

utra

-0.63

orld

-0.63

ocal

-0.62

oard

-0.62

POSITIVE LOGITS

 thresholds

1.02

 threshold

0.93

ansas

0.80

 clearance

0.77

witz

0.75

posts

0.74

 crossings

0.74

 cutoff

0.71

locked

0.70

 heights

0.69

Activations Density 0.015%

terms related to a certain level or limit (a "threshold")

references to quantitative limits or levels of measurement

No Comments

No Known Activations