INDEX

Explanations

verbs related to giving information or informing someone

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "report" and its variations across different contexts

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

creen

-0.82

vous

-0.70

axy

-0.70

å§«

-0.70

egal

-0.69

enta

-0.67

 gauge

-0.66

ortium

-0.66

ucky

-0.64

wich

-0.63

POSITIVE LOGITS

ounces

0.71

 sightings

0.70

Crash

0.68

NX

0.68

 Malone

0.67

ulate

0.66

ribe

0.65

Reporting

0.65

metry

0.64

 reports

0.64

Activations Density 0.060%

verbs related to giving information or informing someone

occurrences of the word "report" and its variations across different contexts

No Comments

No Known Activations