INDEX

Explanations

references to the name "Clare" with varying degrees of importance

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Clare" and related references

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

ĸļ

-0.95

ured

-0.77

HCR

-0.76

ĩ

-0.75

usalem

-0.75

opathy

-0.73

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.72

 Arabia

-0.69

ÑĮ

-0.69

acea

-0.68

POSITIVE LOGITS

mson

0.93

leans

0.92

lic

0.89

lean

0.89

forth

0.88

lli

0.85

fort

0.82

weight

0.81

mont

0.81

nd

0.81

Activations Density 0.042%

references to the name "Clare" with varying degrees of importance

occurrences of the name "Clare" and related references

No Comments

No Known Activations