INDEX

Explanations

reasons or explanations

oai_token-act-pair · gpt-3.5-turbo

reasons or explanations related to the concept of "why."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ymph

-0.68

lator

-0.68

 Roller

-0.68

rop

-0.65

amps

-0.65

aughed

-0.64

phrine

-0.64

shaw

-0.63

Zone

-0.63

ãĤ¹

-0.61

POSITIVE LOGITS

soever

1.00

why

0.99

WHY

0.93

why

0.92

Why

0.75

 exactly

0.75

ihad

0.74

soType

0.69

Origin

0.68

iterranean

0.68

Activations Density 0.031%

reasons or explanations

reasons or explanations related to the concept of "why."

No Comments

No Known Activations