INDEX

Explanations

names of a specific company followed by a numeric value

oai_token-act-pair · gpt-3.5-turbo

occurrences of the abbreviation "AT&T"

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

ãĥ£

-0.76

milo

-0.75

lin

-0.75

 inspir

-0.70

 lessons

-0.67

 powd

-0.65

 Laos

-0.64

dyl

-0.63

 caffe

-0.63

Reviewer

-0.63

POSITIVE LOGITS

ECH

0.99

WA

0.94

ruck

0.89

OWN

0.86

oday

0.84

ributes

0.84

UNE

0.83

CHA

0.83

ronics

0.82

 Mobility

0.82

Activations Density 0.010%

names of a specific company followed by a numeric value

occurrences of the abbreviation "AT&T"

No Comments

No Known Activations