INDEX

Explanations

words related to deception or concealment

oai_token-act-pair · gpt-3.5-turbo

references to "dummy" objects or concepts, particularly in contexts involving concealment or deception

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

ably

-0.76

opher

-0.76

vez

-0.76

ģĸ

-0.74

©

-0.72

zag

-0.69

lez

-0.69

ī

-0.68

 Dame

-0.68

esian

-0.68

POSITIVE LOGITS

 mound

0.75

owship

0.71

cock

0.68

ummy

0.65

 positives

0.64

 habitable

0.63

milo

0.63

ollah

0.62

 depress

0.62

oulos

0.62

Activations Density 0.044%

words related to deception or concealment

references to "dummy" objects or concepts, particularly in contexts involving concealment or deception

No Comments

No Known Activations