INDEX

Explanations

text related to grammar and linguistics

oai_token-act-pair · gpt-3.5-turbo

references to grammar and linguistic concepts

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

urion

-0.90

etheus

-0.80

deen

-0.76

roleum

-0.73

pard

-0.71

ded

-0.70

bledon

-0.70

lein

-0.68

 Volunte

-0.65

ONSORED

-0.65

POSITIVE LOGITS

 diction

1.06

 grammar

1.05

 lingu

1.04

 learners

1.01

 dialect

0.99

 phon

0.97

 lear

0.95

 fluent

0.94

mith

0.93

 proficiency

0.93

Activations Density 0.061%

text related to grammar and linguistics

references to grammar and linguistic concepts

No Comments

No Known Activations