INDEX

Explanations

terms related to publications, publishing, and media

oai_token-act-pair · gpt-3.5-turbo

proper nouns or names related to organizations, titles, and entities in various contexts

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

stellar

-0.73

ĸļ

-0.71

 outgoing

-0.70

 incoming

-0.70

achable

-0.66

 succeeded

-0.65

wikipedia

-0.65

peat

-0.65

 recalling

-0.64

 morphine

-0.64

POSITIVE LOGITS

 Advis

0.84

 Ratings

0.82

 Solutions

0.80

Tracker

0.79

 Partners

0.79

LLC

0.77

 Blog

0.75

 Tracker

0.74

 Finder

0.74

HQ

0.73

Activations Density 0.374%

terms related to publications, publishing, and media

proper nouns or names related to organizations, titles, and entities in various contexts

No Comments

No Known Activations