INDEX

Explanations

verbs with the suffix 'ate'

oai_token-act-pair · gpt-3.5-turbo

words related to negation or invalidation

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

ebus

-0.76

erer

-0.71

ETF

-0.69

ortment

-0.69

acci

-0.67

umbered

-0.66

say

-0.63

erers

-0.63

artney

-0.62

abet

-0.60

POSITIVE LOGITS

 barriers

0.73

 obstacles

0.65

 insur

0.63

 threats

0.63

 virginity

0.63

ãĥķãĤ¡

0.63

 havoc

0.62

 Confederate

0.62

oxide

0.62

 Syndrome

0.61

Activations Density 0.096%

verbs with the suffix 'ate'

words related to negation or invalidation

No Comments

No Known Activations