INDEX

Explanations

messages related to accountability and responsibility towards law enforcement actions

oai_token-act-pair · gpt-3.5-turbo

sections that report or summarize findings and statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 ..........

-0.71

Ãª

-0.68

[+

-0.68

rament

-0.68

atro

-0.68

 Reincarnated

-0.67

 Bagg

-0.67

ibu

-0.66

inventoryQuantity

-0.66

 externalToEVAOnly

-0.66

POSITIVE LOGITS

 summarize

1.01

 summarized

0.99

 informative

0.98

 overarching

0.93

swers

0.89

 summar

0.88

 chronological

0.88

 overview

0.86

 informational

0.82

 anonym

0.82

Activations Density 0.706%

messages related to accountability and responsibility towards law enforcement actions

sections that report or summarize findings and statements

No Comments

No Known Activations