INDEX

Explanations

words related to retractions or retrials

oai_token-act-pair · gpt-3.5-turbo

instances of the word "ret" in various contexts

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

INESS

-0.82

SHIP

-0.76

STEM

-0.74

tips

-0.69

STD

-0.69

¬¼

-0.67

tone

-0.64

Connector

-0.63

hower

-0.62

ppo

-0.62

POSITIVE LOGITS

arations

0.92

ainer

0.91

upt

0.91

ention

0.90

ribut

0.89

itled

0.88

rog

0.88

aped

0.86

raction

0.86

ailed

0.84

Activations Density 0.007%

words related to retractions or retrials

instances of the word "ret" in various contexts

No Comments

No Known Activations