INDEX

Explanations

keywords related to accidents or issues

oai_token-act-pair · gpt-3.5-turbo

specific terms related to incidents, actions, and categorization

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

omever

-0.83

 thereof

-0.76

terday

-0.73

 toget

-0.72

 respectively

-0.70

 acknow

-0.67

olson

-0.67

Downloadha

-0.67

perature

-0.66

âĢ¦."

-0.65

POSITIVE LOGITS

 Profile

0.91

 Releases

0.91

 Description

0.81

 Collection

0.79

 âĵĺ

0.79

 Update

0.79

 Facts

0.79

 Meter

0.77

 Finder

0.77

 Ratings

0.77

Activations Density 0.449%

keywords related to accidents or issues

specific terms related to incidents, actions, and categorization

No Comments

No Known Activations