INDEX

Explanations

instances of phrases or acronyms related to application programming interfaces (APIs)

oai_token-act-pair · gpt-3.5-turbo

references to APIs (Application Programming Interfaces)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

lihood

-0.84

fur

-0.82

interest

-0.81

creen

-0.80

lag

-0.75

ning

-0.75

pard

-0.73

ties

-0.71

wind

-0.69

 Gaia

-0.69

POSITIVE LOGITS

 endpoint

1.01

 Client

0.98

 Token

0.97

 Gateway

0.96

 docs

0.93

API

0.91

api

0.89

 Provider

0.88

 CLIENT

0.88

 wrapper

0.88

Activations Density 0.027%

instances of phrases or acronyms related to application programming interfaces (APIs)

references to APIs (Application Programming Interfaces)

No Comments

No Known Activations