INDEX

Explanations

terms related to awards and recognition

oai_token-act-pair · gpt-3.5-turbo

references to recognitions or awards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

=-=-=-=-

-0.76

roid

-0.70

anza

-0.68

ansas

-0.67

rament

-0.66

liner

-0.66

ration

-0.64

 retri

-0.61

rarily

-0.61

Â¯Â¯Â¯Â¯

-0.61

POSITIVE LOGITS

IBLE

0.87

ISION

0.86

orial

0.83

 honors

0.81

è¯

0.80

ible

0.77

olves

0.76

iate

0.76

ieu

0.76

 Presbyterian

0.75

Activations Density 0.032%

terms related to awards and recognition

references to recognitions or awards

No Comments

No Known Activations