INDEX

Explanations

the word "Good" with emphasis

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of the word "Good," often in relation to various topics or contexts

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

âĹ¼

-0.93

ople

-0.78

ptin

-0.74

-0.70

ADRA

-0.68

pent

-0.67

 externalToEVAOnly

-0.67

 Babel

-0.67

OUP

-0.66

Downloadha

-0.65

POSITIVE LOGITS

bye

1.21

reads

1.06

enough

1.01

 luck

0.98

ness

0.96

luck

0.95

 Samar

0.93

bye

0.90

sword

0.90

 Luck

0.90

Activations Density 0.030%

the word "Good" with emphasis

the repeated mention of the word "Good," often in relation to various topics or contexts

No Comments

No Known Activations