INDEX

Explanations

the word "results"

oai_token-act-pair · gpt-3.5-turbo

mentions of outcomes or findings from studies or research

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

resent

-0.72

 Passage

-0.70

 actionGroup

-0.67

conservancy

-0.66

pton

-0.65

nt

-0.65

aps

-0.63

ModLoader

-0.62

afort

-0.62

ACP

-0.61

POSITIVE LOGITS

iveness

0.85

iments

0.81

results

0.80

ivity

0.79

ively

0.76

 results

0.76

 thereof

0.73

oys

0.73

bringer

0.71

 Results

0.69

Activations Density 0.036%

the word "results"

mentions of outcomes or findings from studies or research

No Comments

No Known Activations