INDEX

Explanations

names starting with "Osa" at different activation levels

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "osa" within words

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

ERAL

-0.83

doms

-0.80

rations

-0.76

sheet

-0.74

rics

-0.74

bler

-0.74

rary

-0.74

rator

-0.70

taking

-0.69

liest

-0.68

POSITIVE LOGITS

osa

1.03

 Luxem

1.02

qua

0.94

que

0.94

velength

0.87

isy

0.84

hea

0.84

Ã±a

0.82

ques

0.81

uce

0.80

Activations Density 0.016%

names starting with "Osa" at different activation levels

occurrences of the substring "osa" within words

No Comments

No Known Activations