INDEX

Explanations

words related to drunkenness

oai_token-act-pair · gpt-3.5-turbo

references to drunkenness and related negative behaviors

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

acea

-0.92

atro

-0.92

ioxide

-0.85

ILA

-0.82

agnetic

-0.78

ä»£

-0.78

ramid

-0.77

irtual

-0.75

isite

-0.74

ordon

-0.74

POSITIVE LOGITS

ness

1.24

ly

1.12

nesses

0.84

 drunken

0.82

liness

0.82

 sailor

0.74

 antics

0.73

 lust

0.72

 disorderly

0.72

 indisc

0.72

Activations Density 0.044%

words related to drunkenness

references to drunkenness and related negative behaviors

No Comments

No Known Activations