INDEX

Explanations

verbs indicating negation or refusal

oai_token-act-pair · gpt-3.5-turbo

verbs and actions related to refusal or non-compliance

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

 largeDownload

-0.65

ortment

-0.63

peria

-0.63

 courtesy

-0.62

ixture

-0.62

itta

-0.61

MC

-0.60

 headed

-0.59

amiliar

-0.58

itton

-0.58

POSITIVE LOGITS

 anymore

1.79

nor

1.32

 anything

1.21

 anywhere

1.15

any

1.15

 anybody

1.12

 whatsoever

1.10

 anytime

1.06

 anyone

1.06

ANY

1.01

Activations Density 0.336%

verbs indicating negation or refusal

verbs and actions related to refusal or non-compliance

No Comments

No Known Activations