INDEX

Explanations

terms related to levels or intensities of a characteristic or attribute

oai_token-act-pair · gpt-3.5-turbo

references to severity or intensity of various situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

icia

-0.78

vl

-0.76

erity

-0.74

frog

-0.73

aughlin

-0.71

BIL

-0.70

ply

-0.70

ISTORY

-0.69

uba

-0.69

mith

-0.69

POSITIVE LOGITS

 underlying

1.00

 respective

0.98

 relationship

0.95

 offending

0.94

 argument

0.93

 spectrum

0.92

 phenomenon

0.89

 latter

0.89

 equation

0.88

 problem

0.87

Activations Density 0.252%

terms related to levels or intensities of a characteristic or attribute

references to severity or intensity of various situations

No Comments

No Known Activations