INDEX

Explanations

adjectives describing manner or condition

oai_token-act-pair · gpt-3.5-turbo

descriptors of conditions and qualities in relation to behavior and interpersonal dynamics

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ammers

-0.80

rists

-0.78

ravings

-0.77

 icons

-0.77

brids

-0.76

 brakes

-0.75

akings

-0.75

itars

-0.74

 tattoos

-0.73

 trophies

-0.72

POSITIVE LOGITS

 manner

1.08

 nutshell

1.00

 context

0.99

 guise

0.89

 vein

0.83

 environment

0.83

 courtroom

0.82

 standpoint

0.81

 fashion

0.80

 sphere

0.80

Activations Density 0.268%

adjectives describing manner or condition

descriptors of conditions and qualities in relation to behavior and interpersonal dynamics

No Comments

No Known Activations